import{B as e}from"./bot-yadxTHRI.js";import{S as t,L as a,a as i}from"./sparkles-BaQiCSRN.js";import{E as r}from"./eye-DT3dI-oM.js";import{C as o}from"./chart-column-BlvAS72H.js";import{C as n}from"./cog-C7yrXwiB.js";import{S as s}from"./server-C1NLTEsX.js";import{S as c}from"./shield-Yh756DzO.js";import{U as l}from"./Footer-BVmbiSy-.js";import{N as d}from"./network-D0ImMo3I.js";import{Z as u}from"./zap-BLOcRQPL.js";const p=[{key:"agentic-ai",slug:"agentic-ai",name:"Agentic AI Systems",promise:"Multi-agent automations with guardrails and HITL review.",bullets:["Eliminate manual back-office steps","Policy-aligned actions","Explainable decisions"],whatYouGet:["Orchestration graph in code (agents, tools, policies) with retries/timeouts","Evals suite: jailbreak, toxicity, groundedness, accuracy gates","Guardrails and safety gates enforced in CI and runtime","Reviewer console (approve/annotate/retry) with audit trail","Run log and trace viewer (inputs, prompts/versions, tool calls, outputs)","Budget caps and alerts; cost per transaction export"],icon:e,gradient:"from-data-orange to-data-amber",outcomes:["Eliminate manual back-office steps","Policy-aligned actions with human override","Traceable, explainable decisions with run logs","Reviewer minutes per case ↓ 30–50%","Cost per transaction capped and reported"],timeline:"2–4 weeks (design+pilot), 6–8 weeks (pilot→prod)",team:"architect, MLE, FE, BE, PM; security reviewer on demand",inputs:["Process maps or workflow documentation","Sample data and edge cases","Policy docs, consent and approval thresholds","SME availability for evaluation","Target KPIs (time saved, accuracy), model/tool constraints"],tech:"Control plane: LangGraph/Temporal/Argo; retries/backoff/idempotency. Models: OpenAI/Claude/local; routing and fallback policies. HITL: reviewer UI; reason codes; audit and retention windows. Identity: SSO (OIDC/SAML); roles/attributes; least-privilege. On-prem/GovCloud or VPC; secrets in KMS; policy repo in Git.",proof:["Before/after KPI chart on target workflow","Evals summary with thresholds and deltas","Run-log sample (redacted) and trace screenshot","Cost per transaction and budget adherence","Policy mapping table (policy → runtime check)","Critical CVEs at release → 0"],faqs:[{question:"What is agentic AI and how does it differ from traditional automation?",answer:"Agentic AI systems make autonomous decisions within defined guardrails, adjusting their approach based on context rather than following rigid if-then rules. Traditional automation executes fixed scripts. Agentic systems evaluate options, chain multi-step actions, and escalate to humans when confidence drops below your threshold. Allerin builds these with policy alignment, explainable decisions, and full run-log traceability."},{question:"How do you prevent agentic AI systems from taking harmful actions in production?",answer:"Three layers. First, policy guardrails define what the agent can and cannot do at the configuration level. Second, human-in-the-loop checkpoints pause execution for review on high-risk actions. Third, real-time monitoring flags anomalous behavior and triggers automatic rollback. Every action is logged with the reasoning chain that produced it."},{question:"How long does it take to deploy an agentic AI system?",answer:"A focused single-workflow agent typically reaches production in 6 to 8 weeks. Multi-agent orchestration across departments takes 12 to 16 weeks. Both timelines include guardrail configuration, evaluation benchmarks, and monitored rollout with rollback capability."},{question:"Can agentic AI integrate with our existing enterprise tools?",answer:"Yes. Allerin builds agents that connect to your existing stack through APIs, webhooks, and database connectors. Common integrations include Salesforce, ServiceNow, Jira, Slack, internal databases, and custom REST endpoints. The agent operates as a new layer on top of your infrastructure, not a replacement."}],relatedProducts:["vista","neurosight"],popularIndustries:["manufacturing","healthcare","finance"],seoTitle:"Agentic AI Systems: Multi-Agent Automation with Guardrails & HITL | Allerin",seoDescription:"Deploy policy-aligned agentic AI automations in 2-4 weeks. Eliminate manual back-office steps with traceable, explainable multi-agent workflows and human oversight.",personas:["CTO/CIO","AI Lead","Operations"],relatedServices:["ai-orchestration","mlops-model-ops","genai-accelerator"],beforeAfterStats:[{metric:"Agent Accuracy",before:"74%",after:"94%",impact:"+20% reduction in hallucinations"},{metric:"Human Review Time",before:"8hr/day",after:"45min/day",impact:"89% efficiency gain"}]},{key:"genai-accelerator",slug:"genai-accelerator",name:"GenAI Product Accelerator",promise:"RAG features shipped to production safely.",bullets:["MVP in weeks","Measurable accuracy","Usage analytics & observability"],whatYouGet:["Vector pipeline + knowledge ingestion (automated re-indexing)","RAG orchestration layer with prompt versioning and fallbacks","Evals suite: accuracy (exact-match + semantic), hallucination gates, toxicity filters","CI/CD integration with regression gates (accuracy thresholds)","Observability dashboard: usage, cost per query, latency p95","Safety monitoring: PII detection, content filters, rate limits"],icon:t,gradient:"from-data-teal to-data-cyan",outcomes:["Working RAG feature in prod with accuracy ≥ target","Evals dashboard and CI check for regressions","Usage analytics and safety monitoring","Measurable user satisfaction or task completion rate","Cost per query optimized and tracked"],timeline:"4 weeks (MVP), 6 weeks (production-ready with full evals and monitoring)",team:"architect, MLE, FE, BE, QA; security reviewer for prod",inputs:["10-50 sample Q&A pairs for evaluation","Source documents or knowledge base","Accuracy targets and success metrics","PII/compliance requirements (HIPAA, SOC2, etc.)","Existing APIs or systems to integrate"],tech:"Vector stores: Pinecone/Weaviate/pgvector with hybrid search. Models: OpenAI (GPT-5/4), Anthropic (Claude 3.5), Google (Gemini), or open-source (Llama 3). Chunking: semantic splitting with overlap; metadata enrichment. Retrieval: BM25 + dense embeddings; reranking with Cohere/cross-encoders. Observability: LangSmith/Phoenix/custom; cost tracking per query. Deployment: Cloud (AWS/GCP/Azure) or on-prem; API Gateway + auth (OAuth2/API keys). Safety: PII redaction, content filters (Azure Content Safety/Llama Guard), rate limits.",proof:["Accuracy scorecard with baseline → production deltas","Hallucination rate chart (pre-launch vs. 30-day avg)","Cost per query breakdown and optimization report","User satisfaction survey results (NPS or task completion)","Retrieval precision/recall metrics by document type","Performance SLA adherence report (latency p50/p95/p99)","Production RAG hallucination rate < 3%"],faqs:[{question:"What is RAG and why does it matter for enterprise GenAI?",answer:"Retrieval-Augmented Generation (RAG) grounds large language model responses in your actual data instead of relying on the model's training data alone. This reduces hallucinations, keeps answers current, and lets you control exactly which documents the model can reference. It is the difference between a chatbot that makes things up and one that cites your internal knowledge base."},{question:"How do you measure RAG accuracy in production?",answer:"We set up eval dashboards that track retrieval precision (did the system find the right documents), answer faithfulness (does the response match the source), and relevance scoring (does the answer address the question). These metrics run continuously against a golden dataset of known-good Q&A pairs specific to your domain."},{question:'What does "production-ready" mean for GenAI features?',answer:"It means the system handles real user traffic with acceptable latency, has safety gates preventing harmful outputs, includes usage analytics for cost monitoring, provides observability into model behavior, and has a tested rollback path. A demo that works on 10 queries is not production-ready. A system that handles 10,000 queries per day with 99.5% uptime is."},{question:"How quickly can we launch a production RAG feature?",answer:"Allerin's GenAI Accelerator delivers a production-ready RAG system in 4 to 6 weeks. That includes data pipeline setup, retrieval tuning, eval framework, safety gates, and deployment with monitoring. The first working prototype is typically ready in week 2."}],relatedProducts:["vista","sentra"],popularIndustries:["retail","healthcare","finance"],seoTitle:"GenAI Product Accelerator: RAG Features Shipped to Production in 4-6 Weeks | Allerin",seoDescription:"Launch production-ready RAG features with measurable accuracy. Includes eval dashboards, safety gates, and usage analytics for secure GenAI deployment.",personas:["Product Lead","Engineering VP","AI Team"],relatedServices:["agentic-ai","computer-vision-fasttrack"],beforeAfterStats:[{metric:"Time to Market",before:"6 months",after:"6 weeks",impact:"4x faster launch"},{metric:"Prompt Accuracy",before:"68%",after:"91%",impact:"+23% user satisfaction"}]},{key:"computer-vision-fasttrack",slug:"computer-vision-fasttrack",name:"Computer Vision FastTrack",promise:"PoC→prod pipeline on edge/cloud.",bullets:["Detect/track reliably","Low latency at the edge","Ops dashboards that stick"],whatYouGet:["Custom model trained on your footage (detection/tracking/classification)","Edge deployment package: ONNX/TensorRT optimized for Jetson/x86/ARM","Inference pipeline with <target latency (typically 60-200ms p95)","Precision/recall benchmarks on test set with confusion matrices","MLOps workflow: drift monitoring, review UI, re-labeling, retraining hooks","Production runbook: deployment, rollback, troubleshooting, scaling"],icon:r,gradient:"from-data-cyan to-data-blue",outcomes:["Model meets precision/recall target on your footage","Edge pipeline ≤ target latency; stable FPS","Ops workflow (review, re-label, retrain) live","Measurable reduction in manual review time (typically 87%)","Drift monitoring and automated retraining triggers"],timeline:"4 weeks (PoC), 8 weeks (production-ready with MLOps)",team:"CV lead, MLE, edge dev, FE, QA; ops reviewer for deployment",inputs:["Sample footage or image sets (200-500 frames minimum)","Labeling guidelines or existing annotations","Target edge hardware specs (Jetson/x86/ARM)","Precision/recall targets and acceptable latency","Ops workflow requirements (drift alerts, retraining triggers)"],tech:"Edge hardware: NVIDIA Jetson (Nano/Xavier/Orin), x86 (Intel/AMD), ARM (RPi/custom). Models: YOLOv8/v11, EfficientDet, custom CNNs; ONNX/TensorRT export; INT8 quantization. Frameworks: PyTorch/TensorFlow → ONNX; TensorRT optimization for 3-5x speedup. Streaming: RTSP/RTMP/USB; frame buffering and batching; resilient reconnect with backoff; timestamp sync with clock drift checks. Deployment: Docker/K3s on edge; REST/MQTT for results; centralized model registry. Edge management: heat throttling guard; GPU temp monitoring; rolling buffers with cold tier (S3/Blob) lifecycle rules. MLOps: Prometheus/Grafana for drift; Label Studio/CVAT for re-labeling; DVC for dataset versioning. Cloud fallback: AWS/GCP/Azure for heavy inference or batch processing.",proof:["Precision/recall scorecard with per-class breakdown","Latency benchmarks (p50/p95/p99) across hardware configs","FPS stability chart (30-day post-deployment)","Drift detection alert examples with remediation","Manual review time reduction (before/after workflow analysis)","Model card with architecture, inputs, outputs, limitations","Production model inference p95 latency ≤ 200ms"],faqs:[{question:"What hardware do you deploy computer vision models on?",answer:"It depends on your latency and cost requirements. NVIDIA Jetson devices (Orin, Xavier) handle most edge deployments where models run on-site. For higher throughput, x86 servers with dedicated GPUs work better. We profile your specific model and data volume to recommend the right hardware before writing any code."},{question:"How do you handle model accuracy drift in production CV systems?",answer:"We build automated drift monitoring that compares live inference distributions against your baseline metrics. When accuracy drops below your threshold, the system alerts your team and can trigger retraining pipelines with newly collected data. This is MLOps for computer vision, not a train-once-and-forget approach."},{question:"What is the typical timeline for a computer vision proof of concept?",answer:"Allerin's FastTrack program delivers a working PoC in 3 to 4 weeks, with the full production deployment completing by week 8. The PoC includes model selection, training on your data, edge optimization, and measured accuracy benchmarks against your success criteria."},{question:"Can you deploy computer vision in environments without reliable internet?",answer:"Yes, that is what edge deployment means. The model runs entirely on local hardware. Inference happens on-device with zero cloud dependency. Results sync to your central system when connectivity is available. This architecture is critical for manufacturing floors, field operations, and locations with air-gapped networks."}],relatedProducts:["vista","sentra","neurosight"],popularIndustries:["manufacturing","warehousing-logistics","retail"],seoTitle:"Computer Vision FastTrack: Edge AI PoC to Production in 4-8 Weeks | Allerin",seoDescription:"Deploy reliable edge computer vision with <target latency, stable FPS. Includes model training, Jetson/x86 deployment, and MLOps workflows for production CV.",personas:["Operations VP","CV Engineer","Plant Manager"],relatedServices:["genai-accelerator","mlops-model-ops"],beforeAfterStats:[{metric:"Model Accuracy",before:"82%",after:"96%",impact:"+14% defect detection"},{metric:"Inference Latency",before:"850ms",after:"120ms",impact:"7x faster processing"},{metric:"Reviewer Minutes/Event",before:"3.8 min",after:"1.9 min",impact:"Target: <2.0 min achieved"}],industryPatterns:[{industry:"Manufacturing",useCase:"Defect detection & assembly verification",example:"Identify surface defects, misalignments, and missing components on production lines with 96%+ accuracy"},{industry:"Transportation & Rail",useCase:"Track/ROW inspection & rolling stock monitoring",example:"Automated detection of rail surface defects, vegetation encroachment, and equipment anomalies reducing manual inspection time by 87%"},{industry:"Warehousing & Logistics",useCase:"Package counting, damage detection & compliance",example:"Real-time package damage detection and volumetric counting with sub-second latency for high-throughput operations"},{industry:"Retail",useCase:"Shelf compliance & queue management",example:"Monitor product placement, planogram compliance, and customer queue depth for operational optimization"}],architectureGuide:[{approach:"Edge-Only",when:"On-prem requirements, no cloud connectivity, <100ms latency needed",tradeoffs:"Lower cost, data sovereignty, limited model size, manual updates",bestFor:"Government/DOT, utilities, remote sites"},{approach:"Hybrid Edge + Cloud",when:"Real-time inference at edge + batch analytics in cloud",tradeoffs:"Best of both worlds, resilient to connectivity loss, moderate complexity",bestFor:"Manufacturing, retail, logistics"},{approach:"Cloud-Fallback",when:"Primary cloud inference with edge backup during outages",tradeoffs:"Simpler edge footprint, higher latency, cloud dependency",bestFor:"Enterprise with reliable connectivity"}],procurementReadiness:["Accuracy thresholds: Precision/recall targets per class with confusion matrices","Latency SLAs: p50/p95/p99 inference latency with hardware specs","Audit trails: Chain-of-custody for training data, model versions, and predictions","Evidence packs: Exportable detection results with timestamps, confidence scores, and visual proof","Retention windows: Configurable data retention policies (RTSP streams, predictions, re-training sets)","Redaction options: PII masking for faces, license plates, and identifiable features","Compliance: CJIS-compliant for law enforcement, SOC 2 Type II, on-prem/GovCloud deployment options"]},{key:"data-analytics-platform",slug:"data-analytics-platform",name:"Data & Analytics Platform",promise:"GIS + KPI dashboards with a data-quality spine.",bullets:["Faster ops insight","DQ pipeline with alerts","Narrative reporting with anomalies"],whatYouGet:["Data connectors with retry logic and monitoring (source → warehouse)","KPI catalog with definitions, owners, refresh schedules, and SLAs","Data quality pipeline: profiling, validation rules, alerts on critical failures","GIS-enabled dashboards with zoom, filter, layer controls, and export","Narrative report generator with automated summaries and anomaly detection","Runbook: troubleshooting, scaling, adding KPIs, data refresh procedures"],icon:o,gradient:"from-data-blue to-primary",outcomes:["Live KPI dashboards with GIS layers and drill-down capabilities","Data quality pipeline with automated alerts on critical field failures","Scheduled narrative reports with anomaly detection and trend analysis","Measurable reduction in report prep time (typically 75%)","Self-service analytics with governed data catalog"],timeline:"3–6 weeks",team:["Data Engineer","Frontend Dashboard Developer","Data Analyst","Project Manager"],inputs:["Data sources and existing workbooks to replace","KPI definitions and data stewardship owners","Data quality target tolerances and thresholds","Security roles and SSO configuration scopes","Reporting cadence and target audiences","GIS layer requirements and spatial data sources"],tech:"Data Warehouses: Snowflake, BigQuery, Redshift, Azure Synapse, Databricks. GIS: ArcGIS Enterprise/Online, QGIS, Mapbox, Google Maps Platform, PostGIS. BI Tools: Tableau, Power BI, Looker, Metabase, custom React dashboards with Recharts/D3. Data Quality: Great Expectations, dbt tests, custom Python validators, Soda. ETL/ELT: Fivetran, Airbyte, dbt, custom Airflow/Prefect DAGs. Orchestration: Airflow, Prefect, dbt Cloud. Integration: REST/GraphQL APIs, CDC (Debezium), message queues (Kafka, RabbitMQ). Observability: Prometheus, Grafana, Monte Carlo, elementary. Export formats: CSV, Excel, PDF, Shapefiles, GeoJSON, ArcGIS Feature Service.",proof:["KPI catalog with owners, definitions, SLAs, and refresh schedules","Data quality scorecard: pass rates, failure alerts, remediation time","Dashboard usage metrics: active users, queries/day, p95 query latency","Before/after report prep time analysis (manual vs automated)","GIS layer performance: render time, zoom responsiveness, concurrent users","Data pipeline SLA tracking: uptime %, late runs, failure recovery time"],faqs:[{question:"What does a production data analytics platform include?",answer:"At minimum: reliable ETL pipelines that keep data fresh, a governed data warehouse with clear ownership, BI dashboards your team actually uses, and self-service query access for analysts. Allerin builds all four layers with data quality checks, access controls, and documentation so the platform does not become another abandoned tool."},{question:"How long does it take to build a data analytics platform?",answer:"A focused MVP covering one business domain (e.g., sales analytics or operational metrics) typically takes 8 to 12 weeks. An enterprise-wide platform with multiple data sources, cross-domain reporting, and self-service access takes 16 to 24 weeks. We scope based on data source complexity and the number of stakeholder groups."},{question:"Can you integrate data from multiple sources into one platform?",answer:"Yes. We build ETL and ELT pipelines connecting databases, SaaS APIs, flat files, streaming sources, and legacy systems into a unified data warehouse. Common sources include Salesforce, HubSpot, Stripe, PostgreSQL, MongoDB, S3, and internal APIs. Data is cleaned, deduplicated, and modeled before reaching your dashboards."},{question:"How do you ensure data quality and governance?",answer:'We implement automated data quality checks at every pipeline stage: schema validation on ingestion, null and duplicate detection during transformation, and freshness monitoring on the warehouse. Governance includes column-level access controls, data lineage tracking, and a business glossary so everyone agrees on what "active customer" means.'}],relatedProducts:["sentra","ipam"],popularIndustries:["manufacturing","warehousing-logistics","transportation-rail"],seoTitle:"Data & Analytics Platform: GIS Dashboards + Data Quality in 3-6 Weeks | Allerin",seoDescription:"Deploy live KPI dashboards with GIS layers, automated data quality pipelines, and narrative reporting. Reduce report prep time by 75% with governed self-service analytics.",personas:["Data Lead","Analytics VP","BI Manager"],relatedServices:["mlops-model-ops","platform-modernization"],beforeAfterStats:[{metric:"Query Performance",before:"45s",after:"1.2s",impact:"97% faster insights"},{metric:"Data Pipeline SLA",before:"82%",after:"99.7%",impact:"Near-zero downtime"},{metric:"Report Prep Time",before:"8 hours",after:"2 hours",impact:"75% time savings"}],industryPatterns:[{industry:"Manufacturing",useCase:"Production KPI dashboards with OEE & yield tracking",example:"Real-time equipment health monitoring with downtime root cause analysis and predictive maintenance alerts integrated with EAM systems"},{industry:"Transportation & Rail",useCase:"Track health GIS with maintenance priority heat maps",example:"Visual inspection data overlaid on GIS with automated work order generation for high-priority segments based on condition scores"},{industry:"Warehousing & Logistics",useCase:"Real-time inventory flow with geofence alerts",example:"Package movement tracking across facilities with dwell-time anomalies and capacity utilization dashboards for operational optimization"},{industry:"Energy & Utilities",useCase:"Asset health dashboards with outage prediction",example:"Grid asset monitoring with failure prediction models, outage impact zones, and crew dispatch optimization via GIS routing"}],architectureGuide:[{approach:"Cloud-Native (Snowflake/BigQuery)",when:"Need scalability, managed services, ML integration, multi-region",tradeoffs:"Best performance and features, cloud costs, requires connectivity",bestFor:"Enterprises with cloud-first strategy, high data volumes"},{approach:"Hybrid (Cloud + On-Prem)",when:"Regulatory constraints, sensitive data on-prem, cloud analytics",tradeoffs:"Balanced compliance and capabilities, moderate complexity",bestFor:"Healthcare, finance, government with data residency requirements"},{approach:"On-Prem Only",when:"Air-gapped environments, full data sovereignty required",tradeoffs:"Complete control, higher ops burden, limited scalability",bestFor:"Defense, critical infrastructure, strict compliance environments"}],procurementReadiness:["Data quality SLAs: Measurable thresholds for completeness, accuracy, timeliness with automated alerts","GIS export formats: Shapefiles, GeoJSON, KML, ArcGIS Feature Service, WMS/WFS endpoints","Dashboard access controls: RBAC, row-level security, SSO/SAML integration, audit logs","Data lineage tracking: End-to-end visibility from source to dashboard with transformation documentation","Performance SLAs: Query latency targets (p95 < 3s), dashboard load time, concurrent user capacity","On-prem/GovCloud deployment: Air-gapped installation support, FedRAMP considerations","Compliance: SOC 2 Type II, HIPAA-ready architecture, data retention policies, PII handling"]},{key:"mlops-model-ops",slug:"mlops-model-ops",name:"MLOps & Model Operations",promise:"Registry, monitoring/drift, and model governance.",bullets:["Traceable models","Drift alerts","Fast rollback"],whatYouGet:["Model registry with versioning, lineage tracking, and metadata tagging (MLflow/W&B/custom)","Drift monitoring dashboard with statistical tests (KL divergence, PSI, data quality)","Automated rollback workflow with last-known-good fallback and rollback criteria","Evaluation harness with precision/recall/F1 benchmarks and confusion matrices","CI/CD integration for model deployment with GitHub Actions/GitLab CI pipelines","MLOps runbook with troubleshooting, scaling guidelines, and cost guardrails"],icon:n,gradient:"from-data-orange to-data-teal",outcomes:["Registry & lineage in place","Drift/accuracy monitoring with alerts","Standardized deploy/rollback flows","Measurable reduction in model deployment time"],timeline:"3–5 weeks",team:["ML Engineer","Platform Engineer","DevOps Lead","QA Engineer"],inputs:["Existing model artifacts and training code","Deployment environments and CI/CD setup","Accuracy thresholds and alert policies","Rollback criteria","Budget/cost guardrails"],tech:"MLflow, Weights & Biases, or custom registry; cloud/on-prem; CI/CD integration (GitHub Actions, GitLab CI)",proof:["Model lineage graph","Drift/accuracy monitoring dashboard","Deployment time before/after","Rollback test evidence"],faqs:[{question:"What is MLOps and why do we need it?",answer:"MLOps is the set of practices that keep machine learning models reliable after deployment. Without it, models degrade silently as data patterns shift. MLOps includes CI/CD pipelines for model updates, drift monitoring, automated retraining triggers, A/B testing for safe rollouts, and performance dashboards. It is the difference between a model that works on launch day and one that works six months later."},{question:"How do you detect when a model's performance is degrading?",answer:"We instrument three monitoring layers. Data drift detection catches when input distributions shift from training data. Prediction drift tracks when model output patterns change unexpectedly. Business metric correlation links model performance to the KPIs you actually care about. Alerts trigger before users notice the problem."},{question:"Can you add MLOps to models we already have in production?",answer:"Yes. Allerin retrofits MLOps infrastructure onto existing deployed models without retraining or redeploying them. We wrap your current models with monitoring, versioning, and rollback capability. Most retrofits take 4 to 6 weeks depending on how many models you're running."},{question:"What MLOps platforms do you work with?",answer:"We build on MLflow, Kubeflow, SageMaker, Vertex AI, and custom Kubernetes-based pipelines depending on your existing cloud infrastructure. The tooling matters less than the practices. We pick the platform that fits your team's skills and your deployment environment."}],relatedProducts:["vista","sentra","neurosight"],popularIndustries:["finance","healthcare","manufacturing"],seoTitle:"MLOps & Model Operations: Drift Monitoring, Registry & Governance | Allerin",seoDescription:"Deploy MLOps infrastructure with model registry, drift monitoring, and automated rollback in 3-5 weeks. Reduce model deployment time with governance.",personas:["ML Platform","DevOps Lead","Data Science VP"],relatedServices:["agentic-ai","data-analytics-platform"],beforeAfterStats:[{metric:"Model Deployment Time",before:"3 weeks",after:"2 hours",impact:"252x faster to production"},{metric:"Drift Detection",before:"Manual/weekly",after:"Auto/real-time",impact:"Proactive model health monitoring"},{metric:"Incident Recovery",before:"4–8 hours",after:"5 minutes",impact:"96% faster recovery with auto-rollback"},{metric:"Canary Failures Caught",before:"Unknown",after:"<1%",impact:"Auto-rollback prevents production incidents"}],industryPatterns:[{industry:"Finance",useCase:"Fraud detection models with strict audit trails",example:"Real-time fraud scoring models with comprehensive lineage tracking, bias monitoring for fair lending compliance, and automated rollback on accuracy degradation below 94%"},{industry:"Healthcare",useCase:"Clinical prediction models with HIPAA compliance",example:"Patient risk stratification models with explainability requirements, PHI-safe model artifacts, audit logs for all predictions, and regulatory-compliant model versioning"},{industry:"Manufacturing",useCase:"Predictive maintenance with edge deployment",example:"Equipment failure prediction models deployed to edge devices with drift monitoring for sensor data, automated retraining triggers, and zero-downtime model updates"}],architectureGuide:[{approach:"Centralized MLOps",when:"Single ML team, consistent tooling requirements, centralized governance needed",tradeoffs:"Strong governance and consistency, but may slow down autonomous teams. Best for orgs prioritizing compliance.",bestFor:"Finance, Healthcare, Regulated Industries"},{approach:"Federated MLOps",when:"Multiple ML teams, poly-cloud deployment, team autonomy prioritized",tradeoffs:"Teams move faster with their preferred tools, but governance becomes harder. Requires strong platform team.",bestFor:"Large enterprises, Multi-cloud, Product-driven orgs"},{approach:"Hybrid MLOps",when:"Balance of governance and flexibility needed, phased adoption",tradeoffs:"Centralize critical governance (registry, monitoring) while allowing tool flexibility. Moderate complexity.",bestFor:"Mid-market, Growing ML teams, Compliance-aware"}],procurementReadiness:["Model lineage and audit trail: Version tracking, dataset provenance, training metadata, deployment history with timestamps","Drift detection SLAs: Real-time monitoring, alert escalation paths, automated rollback thresholds, incident response time commitments","Rollback criteria: Performance degradation thresholds, last-known-good version identification, rollback success rate guarantees","Bias monitoring and fairness: Demographic parity checks, equal opportunity metrics, disparate impact analysis for regulated models","Cost monitoring and guardrails: Training cost budgets, inference cost tracking, automated scale-down policies, cost anomaly alerts","Compliance evidence: SOC 2 Type II attestation, HIPAA compliance for healthcare models, Model risk management (SR 11-7) documentation","Access control and authentication: RBAC for model registry, SSO integration, audit logging for sensitive model access"]},{key:"platform-modernization",slug:"platform-modernization",name:"Platform Modernization",promise:"Modular monolith first; APIs; observability & cost.",bullets:["No feature freeze","p95 down, cost down","Safer releases"],whatYouGet:["Modernization blueprint with target architecture and migration phases","Dual-run infrastructure with blue-green/canary deployment capability","Observability stack: distributed tracing, structured logging, error budgets","API gateway and module boundaries with contracts and versioning","CI/CD pipeline with automated testing and rollback automation","Security baseline: SBOM generation, SAST/DAST scanning, CSP/HSTS headers, secret rotation via KMS","Cost plan: unit economics per request/service, autoscale policies, budget alerts and rightsizing recommendations","Performance report: p95 latency, throughput, cost analysis, error rates"],icon:s,gradient:"from-data-teal to-data-blue",outcomes:["Measurable p95 latency drop & infra cost reduction","Zero-downtime cutover, dual-run validation","Observability + error budget in place","Faster release cadence with lower risk"],timeline:"4–8 weeks",team:["Solution Architect","Backend Engineer","DevOps Lead","QA Engineer"],inputs:["Current architecture docs and pain points","Performance baselines (p95, throughput)","Target cost reduction %","Release cadence goals","Compliance/security requirements"],tech:"Languages: Ruby/Rails, Node.js, Python/Django/Flask, Java/Spring Boot, .NET/C#. Cloud: AWS, Azure, GCP, on-prem/hybrid. Containers: Docker, Kubernetes, ECS, AKS, GKE. CI/CD: GitHub Actions, GitLab CI, Jenkins, CircleCI, Azure DevOps. Observability: Datadog, New Relic, Prometheus/Grafana, Sentry, CloudWatch. Databases: PostgreSQL, MySQL, MongoDB, Redis, DynamoDB. API Gateway: Kong, AWS API Gateway, Azure APIM, Nginx, custom.",proof:["Before/after p95 latency chart","Infra cost comparison","Observability stack (traces, logs, metrics)","Rollback success rate"],faqs:[{question:"How do you migrate a legacy platform without downtime?",answer:'We use a strangler fig pattern: new services run alongside the legacy system, gradually taking over traffic. Critical paths get migrated first with automated rollback. Database migrations happen through dual-write strategies where both old and new systems stay in sync during the transition. Zero downtime means zero downtime, not "just a few minutes."'},{question:"Should we move to microservices or stay monolithic?",answer:"It depends on your team size and deployment frequency. If fewer than 20 engineers deploy weekly, a well-structured modular monolith is usually faster and cheaper to operate. Microservices make sense when independent teams need to deploy different components on different schedules. We assess your situation before recommending either path."},{question:"What cloud providers do you work with?",answer:"AWS, Google Cloud, and Azure. Most modernization projects land on AWS due to service breadth, but we architect for your specific requirements. Some clients use multi-cloud or hybrid setups, which we support with cloud-agnostic infrastructure-as-code using Terraform or Pulumi."},{question:"How long does a typical platform modernization take?",answer:"A focused assessment and migration plan takes 3 to 4 weeks. Actual migration of a mid-complexity platform (monolith to modular architecture, database migration, containerized deployment) takes 12 to 20 weeks depending on the number of integrations and data migration complexity."}],relatedProducts:["ipam","sentra"],popularIndustries:["finance","retail","healthcare"],seoTitle:"Platform Modernization: Zero-Downtime Migration with p95 & Cost Reduction | Allerin",seoDescription:"Modernize legacy platforms without feature freeze. Achieve measurable p95 latency & cost reduction with dual-run validation in 4-8 weeks.",personas:["CTO","Platform Eng","Engineering VP"],relatedServices:["rails-upgrades","security-compliance","mlops-model-ops"],beforeAfterStats:[{metric:"Deployment Frequency",before:"1x/month",after:"12x/day",impact:"360x faster releases"},{metric:"Infrastructure Cost",before:"$85k/mo",after:"$31k/mo",impact:"64% cost reduction"},{metric:"Error Budget Adherence",before:"72% compliance",after:"99.2% compliance",impact:"Near-perfect SLO tracking"}],industryPatterns:[{industry:"Finance",useCase:"Legacy core banking modernization",example:"Migrated COBOL mainframe to Java microservices with zero-downtime, achieving 78% cost reduction and enabling real-time fraud detection"},{industry:"Retail",useCase:"E-commerce platform scalability",example:"Modernized monolithic e-commerce to API-first architecture, handling 12x Black Friday traffic with 82% latency reduction"},{industry:"Healthcare",useCase:"HIPAA-compliant EHR integration",example:"Modernized patient data platform with end-to-end encryption, achieving SOC 2 Type II and 95% faster API response times"}],architectureGuide:[{approach:"Modular Monolith",when:"Teams <50, Rails/Django/monorepo culture, gradual migration preferred",tradeoffs:"Faster deployment (4-6 weeks), simpler ops, but limited independent scaling",bestFor:"SMB, startups scaling to mid-market, risk-averse organizations"},{approach:"Microservices",when:"Large teams (50+), polyglot requirements, independent scaling critical",tradeoffs:"Maximum flexibility and scale, but higher operational complexity and DevOps maturity required",bestFor:"Enterprise, high-growth SaaS, multi-product platforms"},{approach:"Strangler Fig",when:"Legacy systems with high risk, phased migration over 12-24 months",tradeoffs:"Lowest risk with continuous delivery, but longer timeline and dual-system maintenance",bestFor:"Government, heavily regulated industries, mission-critical systems"},{approach:"Hybrid (Modular + Services)",when:"Mid-size teams, balance of governance and flexibility needed",tradeoffs:"Pragmatic approach avoiding microservices sprawl, but requires strong architectural judgment",bestFor:"Growing companies, B2B SaaS, platform teams"}],procurementReadiness:["Performance SLAs: p95 latency targets, throughput guarantees, error budget commitments","Cost projections: Infrastructure cost reduction estimates, ROI timeline, scaling cost models","Zero-downtime guarantee: Dual-run validation, rollback procedures, incident response SLAs","Observability deliverables: Distributed tracing, error budgets, custom dashboards, alert runbooks","Migration evidence: Before/after performance reports, architecture diagrams, cutover checklists","Compliance: SOC 2 Type II, HIPAA, PCI-DSS support for regulated environments","Support: Runbook handoff, 30-day post-launch support, escalation procedures"]},{key:"security-compliance",slug:"security-compliance",name:"AI Security & Compliance",heroHeadline:"97% of Organizations Experienced an AI Security Breach in 2023. Compliance Alone Didn't Stop It.",heroSubheadline:"Your AI systems process sensitive data, make critical decisions, and interact with customers. Traditional compliance frameworks weren't designed for prompt injection, model theft, or data poisoning attacks. You need security built for AI, and compliance that proves it.",promise:"AI-specific security hardening with compliance frameworks that matter.",bullets:["AI threat protection","Continuous monitoring","Audit-ready evidence"],whatYouGet:["AI threat assessment: prompt injection, model theft, data poisoning vulnerability analysis","LLM security hardening: input validation, output filtering, jailbreak protection","Model access controls: RBAC, API key management, usage monitoring","AI-specific SBOM: model lineage, training data provenance, dependency tracking","Compliance mapping: SOC 2, ISO 27001, HIPAA, NIST AI RMF alignment","Continuous monitoring: anomaly detection, drift monitoring, incident response","Audit evidence pack: security controls documentation, penetration test results, compliance artifacts"],icon:c,gradient:"from-data-cyan to-primary",outcomes:["AI-specific vulnerabilities identified and mitigated","Zero critical findings in AI security assessment","SOC 2 + NIST AI RMF compliance evidence delivered","Continuous monitoring and alerting operational","Full audit trail for AI system decisions"],timeline:"4–8 weeks",team:["Security Architect","ML Security Engineer","Compliance Lead","DevSecOps Engineer"],inputs:["AI/ML system architecture documentation","Model inventory and deployment configurations","Current security policies and compliance requirements","Access control matrix and API documentation","Incident response procedures and audit timeline"],tech:"AI security tools (Lakera, Protect AI, Robust Intelligence); SAST/DAST for ML pipelines; model monitoring (Arize, WhyLabs); compliance automation (Vanta, Drata integration); SBOM generation for ML (SPDX, CycloneDX)",proof:["AI threat assessment report with remediation roadmap","Penetration test results for AI endpoints","Compliance mapping to SOC 2, ISO 27001, NIST AI RMF","Model security controls documentation","Continuous monitoring dashboard with alerting","Audit evidence pack with control attestations"],faqs:[{question:"What compliance certifications does Allerin help with?",answer:"CJIS (criminal justice), SOC 2 Type I and II, HIPAA (healthcare), FedRAMP (federal government), and PCI DSS (payment card data). We handle the technical implementation: access controls, encryption, audit logging, vulnerability management, and documentation. We prepare you for the audit, then support you through it."},{question:"How long does it take to achieve SOC 2 compliance?",answer:"SOC 2 Type I (point-in-time assessment) typically takes 8 to 12 weeks from kickoff to audit-ready state. Type II (ongoing compliance over a review period) adds a 3 to 12 month observation window on top of that. The implementation work is the same; Type II just requires sustained evidence collection."},{question:"Can you help with CJIS compliance for law enforcement applications?",answer:"Yes, CJIS is one of our core competencies. We implement the CJIS Security Policy requirements including advanced authentication, encryption at rest and in transit, audit logging, personnel security controls, and incident response procedures. Our team has direct experience with CJIS audits for ALPR and body camera systems."},{question:"Do you provide ongoing compliance monitoring or just the initial setup?",answer:"Both. Initial setup gets you audit-ready. Ongoing monitoring includes continuous vulnerability scanning, access review automation, compliance drift detection, and audit evidence collection. We can run this as a managed service or train your team to operate it independently."}],relatedProducts:["vista","neurosight"],popularIndustries:["finance","healthcare","technology"],seoTitle:"AI Security & Compliance | Secure Your Production AI Systems | Allerin",seoDescription:"97% of organizations experienced an AI security breach in 2023. Protect your AI systems with specialized security assessments, SOC 2 + NIST AI RMF compliance, and continuous monitoring.",personas:["CISO","VP Engineering","Head of AI/ML","Compliance Officer"],relatedServices:["platform-modernization","mlops-model-ops"],beforeAfterStats:[{metric:"Production CVEs",before:"37 critical",after:"0",impact:"100% vulnerability elimination"},{metric:"Audit Prep Time",before:"6 weeks",after:"2 days",impact:"95% faster compliance"},{metric:"SBOM Coverage",before:"0% dependencies tracked",after:"100% supply chain visibility",impact:"Full dependency transparency"},{metric:"Missing Controls Closed",before:"24 ASVS gaps",after:"0 open findings",impact:"Audit-ready control compliance"}],industryPatterns:[{industry:"Healthcare",useCase:"HIPAA compliance for patient data platform",example:"Implemented PHI encryption at rest/transit, BAA templates, audit logging, and breach notification procedures achieving full HIPAA technical safeguards compliance"},{industry:"Finance",useCase:"PCI-DSS compliance for payment processing",example:"Deployed cardholder data tokenization, network segmentation, quarterly ASV scans, and annual penetration tests meeting PCI-DSS Level 1 requirements"},{industry:"Government",useCase:"FedRAMP authorization for cloud services",example:"Delivered SSP documentation, FIPS 140-2 encryption, ConMon integration, and POA&M tracker enabling FedRAMP Moderate ATO"}],procurementReadiness:["Security audit deliverables: OWASP Top 10 report, SAST/DAST findings, penetration test results","Compliance evidence: WCAG 2.1 AA audit report, remediation logs, accessibility test results","SBOM artifacts: CycloneDX/SPDX format, license analysis, vulnerability-to-component mapping","Remediation timeline: Critical CVE fixes (1 week), high-severity (2 weeks), compliance evidence (3-4 weeks)","Framework alignment: OWASP, NIST CSF, CIS Controls, FedRAMP for government contractors","Regulation support: HIPAA (healthcare), SOC 2 Type II (SaaS), PCI-DSS (payments), GDPR (EU data)","Risk documentation: Accepted risk register, compensating controls, remediation roadmap"]},{key:"product-pods",slug:"product-pods",name:"Product Pods",promise:"Dedicated pods + fractional leadership. Predictable throughput, zero-freeze onboarding, audit-ready workflows.",bullets:["Velocity without churn","Transparent cadence","Lower risk"],whatYouGet:["Pod charter with clear ownership, scope boundaries, and escalation paths","Weekly sprint demos with stakeholder Q&A and feedback loops","Bi-weekly retrospectives with actionable improvement items tracked","Monthly roadmap health review: velocity trends, at-risk epics, dependency status","Real-time backlog visibility: definition of ready/done, acceptance criteria templates","Incident response SLOs: P0/P1 triage, post-mortem reports, mitigation tracking","Transparent metrics dashboard: sprint completion, carry-over %, bug density, PR cycle time"],icon:l,gradient:"from-data-blue to-data-orange",outcomes:["Predictable delivery velocity without turnover disruption","Sprint carry-over reduced to <10%","Clear ownership and scope with defined escalation paths","Measurable sprint completion rate >90%","Bug density <2 per 1K LOC; production incidents tracked and closed","Faster time-to-production with continuous delivery cadence"],timeline:"Ongoing engagement; monthly or quarterly renewal",team:["Pod Lead (1 FTE): sprint planning, risk mitigation, stakeholder sync","Frontend Engineers (2 FTE): component library, UI/UX, accessibility, performance","Backend Engineers (2 FTE): API design, data models, integrations, security","QA Engineer (1 FTE): test automation, regression, performance, bug triage","Product Manager (0.5 FTE): feature prioritization, acceptance criteria, metrics"],inputs:["Product roadmap and quarterly priorities with success criteria","Current velocity baseline and key pain points (carry-over, bug rates, lead time)","Definition of Ready (DoR) and Definition of Done (DoD) checklists","Tooling and access: issue tracker, repos, CI/CD, observability, Slack/Teams","Incident and bug SLOs: response time, severity thresholds, escalation policy"],tech:"Stack-agnostic; we adapt to your existing tooling and infrastructure. Common integrations: Jira/Linear, GitHub/GitLab, CircleCI/GitHub Actions, Datadog/Sentry, Slack/Teams. Pod uses your conventions for branching, PR reviews, deployments, monitoring.",proof:["Sprint velocity chart: 12-week trend with baseline and current average","Carry-over trend: % of stories rolled sprint-to-sprint","Roadmap health dashboard: epic progress, at-risk items, dependency blockers","Bug density and incident metrics: critical bugs per release, P0/P1 response times","Sample sprint report: committed vs. completed points, feature breakdown, health scores","Code quality metrics: test coverage %, PR review cycle time, DORA metrics (deployment frequency, lead time)","Exit in a Box sample: knowledge transfer package TOC with runbooks, IaC, docs, and transition timeline"],faqs:[{question:"What is a product pod and how does it differ from staff augmentation?",answer:"A product pod is a dedicated cross-functional team (engineers, QA, DevOps) that takes ownership of product outcomes, not just task completion. Staff augmentation adds bodies to your team. A pod owns a roadmap, runs its own sprints, and is accountable for delivery velocity and quality metrics. You manage the what, the pod manages the how."},{question:"How do you ensure knowledge transfer when a pod engagement ends?",answer:"Knowledge transfer is built into the engagement from day one, not bolted on at the end. The pod writes documentation throughout, pairs with your engineers on critical code paths, maintains runbooks for operational procedures, and conducts structured handoff sessions in the final sprint. We also offer a 30-day support window after handoff."},{question:"What size teams do your product pods include?",answer:"A standard pod is 4 to 6 engineers including a tech lead. Larger initiatives use multiple pods that coordinate through shared architecture standards and integration contracts. Pod composition adapts to your needs: heavier on frontend, backend, data engineering, or DevOps depending on the product requirements."},{question:"How quickly can a product pod start delivering?",answer:"A pod typically reaches productive velocity within 2 to 3 weeks after onboarding to your codebase, tooling, and domain context. First meaningful deliverables land by the end of sprint 2. Full sprint velocity is established by sprint 3 or 4."}],relatedProducts:["ipam","vista","sentra"],popularIndustries:["retail","finance","manufacturing"],seoTitle:"Product Pods: Dedicated Engineering Teams with Fractional Leadership | Allerin",seoDescription:"Scale product velocity without churn. Predictable delivery with dedicated pods, transparent sprint cadence, and measurable roadmap health. 4x faster shipping with <10% carry-over.",personas:["CPO","Engineering VP","Product Lead"],relatedServices:["genai-accelerator","platform-modernization","rapid-prototyping"],beforeAfterStats:[{metric:"Feature Velocity",before:"2.1/sprint",after:"8.4/sprint",impact:"4x faster shipping"},{metric:"Bug Density",before:"12/kloc",after:"1.8/kloc",impact:"85% code quality boost"},{metric:"Carry-Over %",before:"38%",after:"6%",impact:"Cleaner sprint boundaries"},{metric:"Time to Production",before:"3.2 weeks",after:"1.1 weeks",impact:"66% faster shipping"},{metric:"Lead Time for Change",before:"14 days",after:"3 days",impact:"78% faster delivery (p50)"},{metric:"Change Failure Rate",before:"28%",after:"<10%",impact:"Higher release quality"}],industryPatterns:[{industry:"Retail/E-commerce",useCase:"Checkout optimization and inventory sync",example:"Pod reduced cart abandonment by 42% via A/B tested checkout flows and real-time inventory API; shipped 6 payment gateway integrations in 8 weeks."},{industry:"Finance/Fintech",useCase:"Compliance features and fraud detection",example:"Pod delivered SOC 2-ready audit trails, KYC/AML workflows, and ML-based fraud scoring; critical compliance features shipped on regulatory deadlines."},{industry:"SaaS",useCase:"Growth features and onboarding flows",example:"Pod implemented self-serve onboarding with in-app tours, feature flags, and analytics; time-to-value reduced from 7 days to 90 minutes."}],procurementReadiness:["Pod composition and role allocation: detailed breakdown of FTE allocations, responsibilities, reporting structure","Engagement terms: monthly or quarterly renewals, 30-day notice for changes, SLA commitments documented","Deliverables cadence: weekly demos, bi-weekly retrospectives, monthly roadmap health reports with metrics","Pricing transparency: fixed monthly fee per pod, no hidden costs; optional add-ons for specialized roles (ML, DevOps, security)","Handoff and knowledge transfer: documentation, shadowing, training sessions included at no extra cost","Security and compliance: background checks, NDAs, SOC 2/ISO 27001 compliance evidence on request","Tooling and access requirements: issue tracker, repos, CI/CD, observability, Slack/Teams; SSO/SAML support","Incident response commitments: P0/P1 SLOs, escalation paths, post-mortem timelines, on-call rotation if required"]},{key:"ai-orchestration",slug:"ai-orchestration",name:"AI Platform & Orchestration",promise:"One control plane to design, run, observe, and govern multi-agent/LLM apps, with routing, guardrails, and audit trails at enterprise scale.",bullets:["77% lower AI run costs","99% faster model switching","Full audit compliance"],whatYouGet:["3-5 week build","orchestrator + guardrails","run log + observability"],icon:d,gradient:"from-data-orange to-data-amber",outcomes:["Orchestration graph with retries, timeouts, circuit breakers, and backoff logic for multi-agent coordination","Multi-model routing engine with cost/latency policies, automatic fallbacks, and budget enforcement","Central run log with PII redaction, prompt versioning, actor tracing, and compliance-ready audit trails","Guardrails blocking unsafe actions (jailbreak, toxicity, groundedness) with eval suite passing agreed thresholds","Observability stack with traces, metrics, and alerting for latency spikes, cost anomalies, and error rates","SSO/SAML integration, role-based access control (RBAC), and secrets management for secure enterprise deployment"],timeline:"3-5 weeks",team:"architect, MLE, platform eng, QA",inputs:["Agent/tool inventory with dependencies and call graph (e.g., fraud_detector → KYC_agent → document_parser)","Model providers and cost/latency constraints (e.g., GPT-5 for complex reasoning, Llama for high-volume classification)","Safety policies and risk tolerance (jailbreak detection, toxicity thresholds, groundedness checks, PII redaction rules)","Audit and compliance requirements (SOC 2, HIPAA, FedRAMP, GDPR; data residency, retention policies)","Target SLAs (p95 latency <2s, success rate >99%, budget cap $X/month, uptime 99.9%)"],tech:"LangGraph/Temporal/Argo for orchestration; multi-model routing (OpenAI GPT-5/4, Anthropic Claude, Google Gemini, Llama/Mistral); vector stores (Pinecone, Weaviate, pgvector); observability (Datadog/Grafana, OpenTelemetry); on-prem/GovCloud deployment; SSO/SAML (Okta, Auth0); secrets management (Vault, AWS Secrets Manager)",proof:["SLA verification: p95 latency and success-rate targets met on 3 critical flows with load testing results","Cost governance: Budget caps enforced with real-time alerts; model switchovers proven with A/B test logs","Audit trail: Run log traces every request with actor ID, timestamp, prompt version, tool calls, outputs, and PII redaction proof","Safety validation: Guardrails block unsafe actions in test scenarios; eval suite passes agreed accuracy/safety thresholds","Orchestration graph sample: DAG visualization with retry logic, timeouts, and agent dependencies (available on request)","Cost optimization report: Model routing savings breakdown with before/after spend analysis (available on request)"],faqs:[{question:"What is AI orchestration and why do enterprises need it?",answer:"AI orchestration is a centralized control layer that manages routing, cost, and governance across multiple AI models and providers. Without it, teams independently spin up OpenAI, Anthropic, and open-source models with no visibility into total spend, no consistent safety policies, and no way to compare performance. Orchestration gives you one place to manage all of it."},{question:"How does multi-model routing work?",answer:"Requests are routed to different models based on rules you define: complexity, cost sensitivity, latency requirements, or data sensitivity. Simple queries go to cheaper, faster models. Complex reasoning tasks go to more capable models. Sensitive data stays on private models. The routing logic is configurable and auditable."},{question:"Can AI orchestration reduce our model costs?",answer:"Typically by 30 to 60 percent. Most enterprises are sending every request to their most expensive model because it is the only one configured. Smart routing sends 70 to 80 percent of requests to cheaper models that handle them just as well, reserving expensive models for the queries that actually need them. Budget caps prevent cost overruns."},{question:"Does AI orchestration work with our existing AI applications?",answer:"Yes. The orchestration layer sits between your applications and the model providers. Your apps send requests to the orchestrator instead of directly to OpenAI or Anthropic. The switch typically requires changing one API endpoint per application. No model retraining, no application rewrites."}],relatedProducts:["vista","neurosight"],popularIndustries:["finance","healthcare","manufacturing"],seoTitle:"AI Orchestration Platform: Multi-Agent Control Plane for Enterprise | Allerin",seoDescription:"Enterprise AI orchestration with 77% cost reduction, 99% faster model routing, and full audit compliance. Multi-agent coordination, guardrails, and observability deployed in 3-5 weeks.",personas:["AI Platform Lead","CTO","Engineering VP"],relatedServices:["agentic-ai","mlops-model-ops","genai-accelerator"],beforeAfterStats:[{metric:"Model Switching Time",before:"3 weeks",after:"5 minutes",impact:"99% faster routing"},{metric:"AI Run Costs",before:"$18k/mo",after:"$4.2k/mo",impact:"77% cost reduction"},{metric:"Agent Coordination Latency",before:"8.4s",after:"1.2s",impact:"86% faster multi-agent flows"},{metric:"API Call Volume",before:"2.4M/mo",after:"890k/mo",impact:"63% fewer redundant calls"},{metric:"Audit Compliance Score",before:"62%",after:"98%",impact:"Full SOC 2 readiness"},{metric:"Model Fallback Success",before:"73%",after:"99.2%",impact:"Zero downtime on provider outages"}],industryPatterns:[{industry:"Finance",useCase:"Multi-agent fraud detection with real-time coordination",example:"Global bank deployed 8-agent fraud detection system: transaction_analyzer → risk_scorer → KYC_validator → document_verifier → decision_engine. Orchestration reduced false positives by 41% via intelligent model routing (Llama for fast scoring, GPT-5 for complex case review) and cut per-transaction cost from $0.18 to $0.04. Audit log provides full trace for regulatory compliance (FINRA, FinCEN)."},{industry:"Healthcare",useCase:"Clinical workflow automation with HIPAA-compliant orchestration",example:"Healthcare system automated prior authorization with 5-agent workflow: intake_bot → clinical_reviewer → policy_checker → approval_engine → notification_sender. Orchestration enforced PII redaction, provided HIPAA audit trails, and reduced authorization time from 4.2 days to 6 hours. On-prem deployment with SSO integration for 1,200 clinicians."},{industry:"Manufacturing",useCase:"Supply chain optimization with agentic planning",example:"Manufacturer deployed 6-agent supply chain planner: demand_forecaster → inventory_optimizer → supplier_coordinator → logistics_router → risk_assessor → decision_reporter. Orchestration reduced planning cycle from 3 days to 4 hours, cut expedited shipping by 38%, and provided real-time visibility into agent decision logic for executive reporting."}],procurementReadiness:["Pricing tiers: (1) Per-agent: $2,500-$5,000/agent/month for 1-5 agents; (2) Platform fee: $15k-$30k/month for 6-20 agents with observability and governance; (3) Custom enterprise: Quote for 20+ agents with SLA commitments and dedicated support","SLA commitments: 99.9% uptime, p95 latency <2s, success rate >99%, 1-hour response for P0 incidents, 4-hour response for P1; post-mortem reports within 48 hours","Security and compliance: SOC 2 Type II, HIPAA, FedRAMP, ISO 27001 certifications; PII redaction, data residency controls, encryption at rest and in transit; compliance evidence package available on request","Deployment options: SaaS (multi-tenant), dedicated VPC (single-tenant), on-prem (your data center), GovCloud/Azure Government; SSO/SAML integration (Okta, Auth0, Azure AD)","Deliverables: Orchestration platform with DAG editor, run log dashboard, cost tracking, guardrails config, observability stack; technical documentation, runbooks, training sessions; 30-day post-launch support","Observability and monitoring: Real-time dashboards for latency, cost, error rates; alerting for budget spikes, SLA violations, guardrail triggers; OpenTelemetry traces exported to your stack (Datadog, Grafana, Splunk)","Model provider flexibility: Bring your own API keys (OpenAI, Anthropic, Google) or use our managed endpoints; support for local models (Llama, Mistral) and custom fine-tuned models; no vendor lock-in","Knowledge transfer and handoff: Full source code, IaC (Terraform/CloudFormation), CI/CD pipelines, runbooks, and architecture diagrams; 2-week shadowing period with your team; optional ongoing managed services"],acceptanceCriteria:["Orchestration graph deployed with retries, timeouts, circuit breakers, and backoff logic for multi-agent coordination","Multi-model routing engine configured with cost/latency policies, automatic fallbacks, and budget enforcement tested on 3 critical flows","Central run log capturing 100% of requests with actor ID, timestamp, prompt version, tool calls, outputs, and PII redaction validated","Guardrails blocking unsafe actions (jailbreak, toxicity, groundedness) with eval suite passing agreed accuracy thresholds (e.g., >95% precision)","Observability stack (Datadog/Grafana) integrated with traces, metrics, alerting for latency spikes, cost anomalies, and error rates","SSO/SAML authentication configured with RBAC for platform access; secrets management (Vault/AWS Secrets Manager) integrated","SLA targets met on load testing: p95 latency <2s, success rate >99%, cost per request within budget cap","Compliance requirements validated: Audit logs meet SOC 2/HIPAA/FedRAMP standards; data residency and retention policies enforced","Knowledge transfer completed: Technical documentation, runbooks, IaC, and 2-week shadowing period with your team","Post-launch support: 30-day on-call coverage with 1-hour P0 response, 4-hour P1 response; post-mortem process documented"]},{key:"rapid-prototyping",slug:"rapid-prototyping",name:"Rapid Prototyping Lab",promise:"Executive-ready prototypes + tech spikes in 2 weeks. De-risk before you build: clear go/no-go with user validation and feasibility proof.",bullets:["Stakeholder buy-in","Validated UX + tech spike","Clear build/no-build call"],whatYouGet:["2-week lab","user-tested prototype","tech spike"],icon:a,gradient:"from-data-cyan to-data-teal",outcomes:["Clickable prototype covering 5-7 critical flows","Tech spike code for riskiest section (e.g., RAG, CV, SSO)","Risk register with cost/effort T-shirt sizing","Clear decision: proceed, park, or pivot"],timeline:"2 weeks",team:"architect, UX, FE, MLE/BE (spike)",inputs:["User journeys or workflow maps","Riskiest technical assumption (RAG accuracy, latency, integration)","Target users for testing (5+ participants)","Decision criteria and stakeholders","Budget and timeline constraints"],tech:"Figma or code-based prototype; tech spike in target stack (Python/Node/Rails); cloud or on-prem",proof:["User testing session recordings + synthesis","Tech spike benchmarks (latency, accuracy, feasibility)","Risk register with mitigation options","Decision brief with next-step recommendations"],faqs:[{question:"What do you deliver in a 2 to 3 week rapid prototype?",answer:"A working functional prototype that demonstrates the core AI capability against your real data. Not wireframes, not slide decks. You get a testable application with measured accuracy, latency benchmarks, and a clear go/no-go recommendation. The prototype includes enough infrastructure to run user tests with actual stakeholders."},{question:"What happens after the prototype phase?",answer:"Three possible outcomes. If the prototype validates the concept, we scope a production buildout with realistic timelines. If it partially validates, we adjust the approach and run another focused iteration. If it fails to meet your criteria, you have a clear technical analysis explaining why, which saves you months of building the wrong thing."},{question:"Do we need to provide training data for the prototype?",answer:"Ideally yes, but we can work with limited data. If you have existing datasets, we use those. If not, we can generate synthetic data for initial validation or use publicly available datasets that approximate your domain. The prototype's accuracy improves significantly with real data, which is why we recommend providing it if possible."},{question:"How is rapid prototyping priced?",answer:"Fixed-scope engagement with a defined deliverable and timeline. You know the cost before we start. No hourly billing surprises, no open-ended discovery phases. The prototype deliverable and success criteria are agreed upon in the first two days."}],relatedProducts:["vista","neurosight"],popularIndustries:["manufacturing","retail","finance"],seoTitle:"Rapid Prototyping Lab: 2-Week Executive Prototypes & Tech Spikes | Allerin",seoDescription:"De-risk product decisions with user-tested prototypes and technical feasibility spikes in 2 weeks. Clear build/no-build recommendations with risk analysis.",personas:["CTO/CIO","Product Lead","Innovation"],relatedServices:["genai-accelerator","computer-vision-fasttrack","product-pods"],beforeAfterStats:[{metric:"Decision Cycle",before:"6 weeks",after:"2 weeks",impact:"3x faster validation"},{metric:"Wasted Dev Time",before:"4 sprints",after:"0",impact:"De-risked before commit"},{metric:"Cost of Wrong Decision",before:"$400k wasted",after:"$18k prototype",impact:"96% cost avoidance"},{metric:"Stakeholder Alignment",before:"8 weeks",after:"2 weeks",impact:"75% faster buy-in"},{metric:"Feature Validation Rate",before:"40% unused",after:"85% adopted",impact:"2x feature success"},{metric:"Risk Identification",before:"20% risks found",after:"90% risks mapped",impact:"4.5x risk visibility"}],industryPatterns:[{industry:"Manufacturing",useCase:"CV-based defect detection feasibility",example:"Prototyped defect dashboard + spike on model accuracy for top 3 defect types. Validated 91% accuracy, greenlit with $1.2M/yr savings projection."},{industry:"Retail",useCase:"GenAI personalization validation",example:"Clickable e-commerce flow + RAG spike with product catalog. User testing revealed checkout friction; pivoted design, de-risked integration."},{industry:"Finance",useCase:"Fraud detection ML + compliance",example:"Tech spike on model precision/recall + compliance framework prototype. Proved 94% precision, mapped audit trail requirements, greenlit for build."}]},{key:"integration-fasttrack",slug:"integration-fasttrack",name:"Integration FastTrack",promise:"Production-ready connectors with test suites and docs.",bullets:["Zero-surprise cutovers","Fewer data errors","Battle-tested retries"],badges:["Zero-Downtime","Contract-Driven","Dual-Run Validation","Production-Ready"],whatYouGet:["Spec pack: sequence diagrams, auth scopes, OpenAPI/Protobuf/Avro/JSON schema","Test harness: unit/contract/integration; replayable fixtures; mock servers","Error taxonomy + retry/backoff and idempotency keys","Reconciliation jobs + variance dashboards; backfill plan","Runbooks: deploy/rollback, replay, DLQ handling, on-call workflow","Security: SAST/DAST/SCA clean; secrets in KMS; PII masking; least-privilege","Docs: setup, versioning policy, change log, support SLAs"],icon:i,gradient:"from-data-blue to-data-cyan",outcomes:["≥99.5% successful sync on sample data","Webhook + retry/idempotency logic battle-tested","Dual-run reconciliation with <0.5% variance","Alerting, runbooks, and versioned docs"],timeline:"3–6 weeks depending on protocol count and vendor responsiveness",team:"Integration lead, BE, BE/SRE, QA, architect; optional security reviewer",inputs:["Source/target contracts (OpenAPI/GraphQL/EDIFACT/X12/FHIR/NIEM/etc.)","Test credentials + sandbox tenants; sample datasets; data ownership","Error/timeout/rate-limit policy; throughput targets; idempotency rules","Mapping specs; transformation rules; PII handling","Environments, CI/CD, and observability access"],tech:"Middleware: MuleSoft, Boomi, Apache Camel, Node.js/Python. Message Brokers: Kafka, RabbitMQ, AWS SQS/SNS. API Management: Kong, Apigee, AWS API Gateway. Contract Testing: Pact, Postman/Newman. Monitoring: Datadog, Prometheus + Grafana.",proof:["Sync success rate dashboard (≥99.5%)","Dual-run reconciliation report","Contract test suite and fixtures","Runbook and alerting config"],faqs:[{question:"What types of system integrations does Allerin handle?",answer:"API-to-API integrations between SaaS platforms, database synchronization across systems, legacy system connectors using custom middleware, event-driven integrations via webhooks and message queues, and file-based integrations for systems that only support batch exports. We connect anything that has an interface, documented or not."},{question:"How do you handle integrations with legacy systems that lack APIs?",answer:"We build custom middleware that translates between modern API protocols and whatever the legacy system supports: database direct connections, file drops, screen scraping for systems with only a UI, or custom socket protocols. The middleware presents a clean API to your modern systems while handling the legacy complexity behind the scenes."},{question:"What happens when an integration breaks in production?",answer:"Every integration we build includes monitoring, alerting, and retry logic. Failed messages go to a dead letter queue with full context for debugging. The system alerts your team within minutes of a failure, not when a user reports missing data. We also build circuit breakers that prevent cascade failures across connected systems."},{question:"How long does a typical integration project take?",answer:"A standard two-system API integration takes 2 to 4 weeks. Multi-system integrations with complex data transformations take 6 to 10 weeks. Legacy system connectors with custom middleware add 2 to 3 weeks depending on the legacy system's interface complexity."}],relatedProducts:["ipam","sentra"],popularIndustries:["manufacturing","retail","government","energy-utilities"],seoTitle:"Integration FastTrack: Enterprise System Integration in 8-12 Weeks | Allerin",seoDescription:"Bridge legacy systems to modern platforms with 99.7% sync success. Includes dual-run validation, contract testing, and zero-downtime deployment for ERP, WMS, CAD/RMS, and GIS systems.",personas:["Integration Eng","Platform Eng","CTO"],relatedServices:["data-analytics-platform","platform-modernization"],beforeAfterStats:[{metric:"Sync Success Rate",before:"91.8%",after:"99.7%",impact:"+7.9% reliability"},{metric:"Integration Time",before:"18 weeks",after:"10 weeks",impact:"44% faster delivery"},{metric:"Data Variance",before:"2.3%",after:"0.3%",impact:"87% improvement"}],industryPatterns:[{industry:"Manufacturing",useCase:"ERP ↔ MES ↔ QMS Integration",example:"Real-time production data sync with <500ms latency. 99.7% sync success across 800K+ records. Zero production impact during cutover."},{industry:"Retail & E-Commerce",useCase:"E-commerce ↔ ERP ↔ WMS ↔ Payment Integration",example:"Order-to-cash flow integrity with <1 second latency. Inventory variance <0.1%. Seamless Black Friday 10x load handling."},{industry:"Government & Public Sector",useCase:"CAD/RMS ↔ Records Management ↔ GIS Integration",example:"CJIS-compliant criminal justice data integration. 100% audit trail coverage. Zero data loss during cutover."},{industry:"Energy & Utilities",useCase:"SCADA ↔ GIS ↔ Asset Management Integration",example:"Real-time asset status sync. GIS layer reconciliation <0.5% variance. Outage event propagation <2 minutes."}],architectureGuide:[{approach:"Strangler Fig Pattern",when:"Legacy modernization with gradual migration",tradeoffs:"Lower risk, longer timeline, dual-run complexity",bestFor:"ERP migrations, mission-critical systems"},{approach:"Event-Driven Integration",when:"Real-time sync requirements",tradeoffs:"Scalable, resilient, requires message broker infrastructure",bestFor:"E-commerce, IoT, high-volume transactions"},{approach:"API Gateway + BFF",when:"Multiple consumers (mobile, web, partners)",tradeoffs:"Centralized control, potential bottleneck",bestFor:"Multi-channel applications, API economy"}],procurementReadiness:["All data classified per your schema (PII, PHI, CUI, public)","Encryption in transit (TLS 1.3) and at rest (AES-256)","Audit logging for every API call and data access event","Open standards preferred (REST/JSON, OpenAPI) over proprietary formats","Source code and IP ownership transferred under work-for-hire agreement","HIPAA BAA, PCI-DSS, SOC 2 Type II, CJIS compliance available","Fixed-price engagement with milestone-based payments (30/40/30)","SLA: 99.9% uptime, <1 hour P1 response, 30 days post-production support"],compatibilityMatrix:{protocols:["REST/GraphQL/gRPC","SOAP","SFTP/S3","Kafka/Kinesis/PubSub","Webhooks"],messageBrokers:["Kafka","RabbitMQ","AWS SQS/SNS","Azure Service Bus","Google Pub/Sub"],messageFormats:["JSON/Avro/Protobuf/NDJSON/CSV","EDI X12 (270/271/276/277/835/837)","HL7/FHIR"],targetSystems:["SAP/SuccessFactors","Oracle","Workday","Salesforce","ServiceNow","Dynamics","Netsuite","Okta/Azure AD (SCIM/OIDC/SAML)","Snowflake/BigQuery/Redshift","ArcGIS/Feature Services","S3/GCS/Azure Blob","Triton/MLflow","VMS/WMS/TMS/EAM/CMMS"]}},{key:"rails-upgrades",slug:"rails-upgrades",name:"Rails Upgrades without Feature Freeze",heroHeadline:"Ruby on Rails Upgrades That Don't Break Your Business",heroSubheadline:"Zero-downtime migrations from legacy Rails versions to modern, secure, high-performance applications. Trusted by startups, enterprises, and everything in between.",promise:"In-place Rails/Ruby upgrades while you keep shipping. Dual-boot + blue/green, strong_migrations, CI hardening, and SLO-driven tuning.",bullets:["Same-day cutovers, no freeze","p95 down ≥ 30%","Zero critical CVEs"],badges:["Dual-boot","Blue/green","strong_migrations","SLO-driven"],trustIndicators:["No credit card required","Assessment in 5 business days","100% confidential"],whatYouGet:["Upgrade plan: Ruby X→Y, Rails A→B; gem audit and shim strategy","Dual-boot enabled; green path proven in staging with traffic replay","Zero-downtime migrations via strong_migrations / gh-ost / pt-osc; backout path","CI/CD hardening: matrix builds (old/new), contract tests, flaky-test quarantine","Observability pack: pre/post p95, p99, error budgets, Slow Query log reports","Performance fixes: N+1 elimination; index strategy, partitioning; cache keys","Security hardening: Brakeman, bundler-audit/Snyk, CSP/HSTS, CSRF, session store, key rotation","Cost controls: puma worker math, pgbouncer, env-specific pool sizes, object store offload","Cutover runbook: canary %, health gates, 'abort switch,' rollback <10 min","Post-go-live hypercare (2-4 weeks) with SLO watch"],icon:u,gradient:"from-data-amber to-data-teal",outcomes:["Same-day cutovers with dual-boot and safe migrations (no freeze)","Zero critical CVEs at release; dependency policy enforced in CI","p95 down ≥ 30% on named hot paths; error rate not worse","Deploy frequency ≥ daily with automatic canary and rollback","Infra and DB cost reduced 20-40% with YJIT, caching, pool tuning"],timeline:"4-8 weeks",team:"Rails lead, BE, SRE/DevOps, QA, Sec reviewer",inputs:["Codebase + Gemfile.lock, production configs, deploy scripts","Current perf/security reports; SLOs; infra costs; DB stats (pg_stat_statements)","Release calendar and downtime constraints","List of critical user journeys and SLAs","Access to CI/CD, observability, error tracking, DB consoles"],tech:"Ruby 3.2/3.3 with YJIT; Rails 7.x (Zeitwerk); Puma 6; PostgreSQL ≥13; strong_migrations; pgbouncer; OpenTelemetry; Datadog/New Relic/Grafana",proof:["All tests passing in both Rails versions","CVE count → 0 at release; SBOM generated","p95 latency improved ≥30% on defined endpoints","Rollback proven <10 minutes; blue-green cutover report"],faqs:[{question:"Can you upgrade Rails without taking our application offline?",answer:"Yes, zero-downtime upgrades are our standard approach. We use blue-green deployments or rolling updates so the old version handles traffic while the new version starts up. Database migrations are written to be backward-compatible so both versions can run simultaneously during the cutover window."},{question:"How do you handle gem compatibility issues during Rails upgrades?",answer:"We audit every gem in your Gemfile for version compatibility with the target Rails version before starting. Gems with breaking changes get updated or replaced with maintained alternatives. Abandoned gems get forked and patched if no replacement exists. The dependency resolution happens upfront so there are no surprises mid-upgrade."},{question:"What Rails versions do you upgrade from and to?",answer:"We handle upgrades from Rails 4.2 through Rails 7+, including multi-version jumps. We upgrade one major version at a time (4 to 5, then 5 to 6, then 6 to 7) because skipping versions introduces compounding compatibility issues. Each version increment is a tested, deployable milestone."},{question:"Do you also handle Ruby version upgrades?",answer:"Yes. Rails upgrades often require Ruby version bumps, and we handle both together. Ruby version changes affect all gems in your application, not just Rails, so we test the full dependency tree. We also address any Ruby syntax deprecations and performance improvements available in the newer version."}],relatedProducts:["ipam"],popularIndustries:["finance","retail","healthcare"],seoTitle:"Rails Upgrades without Feature Freeze: dual-boot, blue-green, measurable p95 & cost wins | Allerin",seoDescription:"In-place Rails/Ruby upgrades while you keep shipping. Dual-boot + blue/green, strong_migrations, CI hardening, and SLO-driven tuning. 4–8 week sprint with audit-ready evidence.",personas:["Rails Lead","Backend Eng","CTO"],relatedServices:["platform-modernization","security-compliance"],beforeAfterStats:[{metric:"p95 latency",before:"2.4s",after:"280ms",impact:"target ≥ 30-60% drop"},{metric:"Deploy frequency",before:"1/month",after:"12/day",impact:"continuous delivery enabled"},{metric:"Infra cost",before:"$85k/mo",after:"$31k/mo",impact:"right-size + caching + YJIT"},{metric:"CVEs at release",before:">0",after:"0",impact:"Brakeman/bundler-audit/SBOM clean"}],industryPatterns:[{industry:"Finance & Fintech",useCase:"SaaS Platform Upgrade (Rails 6.1 → 7.1)",example:"Upgraded fintech API serving 2.4M transactions/day. Zero downtime, p95 from 1.8s → 680ms. PCI-DSS compliance maintained throughout. 18 CVEs eliminated."},{industry:"Retail & E-Commerce",useCase:"E-commerce Monolith Upgrade (Rails 5.2 → 7.1)",example:"Black Friday-ready Rails upgrade for 800K SKU catalog. 55% memory reduction enabled downsizing from 24 to 14 dynos. $86k annual savings."},{industry:"Healthcare & MedTech",useCase:"HIPAA-Compliant Patient Portal (Rails 4.2 → 7.1)",example:"Multi-version Rails upgrade with audit trail preservation. BAA compliance maintained. Zero data loss during dual-boot migration."}],architectureGuide:[{approach:"Dual-Boot",when:"Monolith with active feature work",tradeoffs:"Longer timeline (6-8 weeks), more CI complexity, but zero feature freeze",bestFor:"Teams shipping 20+ PRs/week, high-traffic apps, SaaS platforms"},{approach:"Blue-Green",when:"Clean cutover, strict rollback SLA (<10 min)",tradeoffs:"Requires 2x infrastructure temporarily, simpler CI setup",bestFor:"High-availability systems, financial services, enterprise SaaS"},{approach:"In-Place + Canary",when:"Small teams, lower traffic, faster timeline",tradeoffs:"Higher risk, requires robust monitoring and alerting",bestFor:"Startups, MVPs, low-complexity apps (<50K LOC)"}],procurementReadiness:["Source code and test suite audit with deprecation inventory","Zero-downtime deployment with rollback SLA (<10 min)","All tests passing in both Rails versions before cutover","CVE remediation with SBOM (CycloneDX format)","Performance SLA: ≥30% p95 improvement on defined endpoints","IP ownership: upgraded codebase and gem shims under work-for-hire","Fixed-price with milestone payments (30/40/30)","30 days post-cutover support included"],compatibilityMatrix:{protocols:["Rails 4.x → 7.x","Rails 5.x → 7.x","Rails 6.x → 7.x","Ruby 2.7 → 3.3"],messageBrokers:["Sidekiq","Resque","DelayedJob","ActiveJob"],messageFormats:["Heroku","AWS (ECS/EC2)","Render","Fly.io","Bare Metal","Docker/K8s"],targetSystems:["PostgreSQL","MySQL/MariaDB","Redis","Memcached","Devise","Pundit","CanCanCan","ActiveAdmin","RSpec","Minitest","Webpacker → Vite","Sprockets → Propshaft"]}}],m=e=>p.find(t=>t.slug===e);export{m as g,p as s};