Sr. Developer (Python/FastAPI) 5+ Years
No. of positions - 3
Job Type - Full Time Employment
Mode of work : WFO
Shift timings : 2PM/3PM - Flexibility is required per the business need
Position Overview :
We need three Senior Backend Developers to build the core microservices layer that powers the conversation engine, voice processing pipeline, risk alerting system, and clinical integrations. A HIPAA-compliant, voice-first AI therapeutic platform for adults recovering from Substance Use Disorder (SUD). The system features a local AI agent architecture with zero external API dependencies — all LLM inference, speech processing, and data storage runs on self-hosted GPU infrastructure.
Key Responsibilities
- Design and develop production-grade microservices using Python (FastAPI) with async support for real-time voice conversation workflows
- Build the Conversation Orchestration Service — dialogue state management, context memory (Redis), NLU routing, and response generation pipeline integrating with local Llama 3.1 8B LLM via vLLM inference server
- Develop the Voice Services layer — WebSocket streaming interface between mobile apps and local Whisper ASR + StyleTTS2 TTS models, audio buffering, turn-taking protocol, and latency optimization (<1s target)
- Implement the Risk & Alerts Service — rule engine for multi-tier escalation (Critical/Urgent/Warning), real-time alert routing via RabbitMQ/Redis Streams, and SMS/push notification triggers
- Build the Integration Hub — HL7 FHIR R4 bi-directional EHR integration, wearable data ingestion (HealthKit/Google Fit), telehealth connectors, and crisis hotline routing
- Implement API gateway patterns using Nginx/Kong (self-hosted) with rate limiting, request validation, and routing
- Develop authentication and authorization flows using Keycloak (self-hosted) — OAuth 2.0, JWT, MFA, RBAC with role-based access for patients, providers, and admins
- Build data access layers for PostgreSQL (clinical data), MongoDB (conversation logs), TimescaleDB (time-series), and Redis (session state)
- Implement HIPAA and 42 CFR Part 2 compliance controls at the application layer — audit logging, consent management, PHI redaction, and redisclosure prevention
- Write comprehensive unit and integration tests; participate in code reviews; maintain API documentation
Required Qualifications
• 5+ years of professional Python backend development experience
- Strong proficiency with FastAPI or equivalent async Python frameworks (Starlette, Sanic)
- Experience building real-time streaming systems using WebSockets or gRPC
- Hands-on experience with PostgreSQL, MongoDB, and Redis in production
- Solid understanding of microservices architecture, event-driven systems, and message queues (RabbitMQ, Redis Streams, or Kafka)
- Experience with Docker containerization and orchestration (Docker Compose, K3s, or Kubernetes)
- Strong understanding of OAuth 2.0, JWT-based authentication, and RBAC patterns
- Experience writing comprehensive test suites (pytest, integration testing)
- Familiarity with CI/CD pipelines (GitLab CI, GitHub Actions)
Preferred Qualifications
- Prior experience in healthcare software development (HIPAA, HL7 FHIR, 42 CFR Part 2)
- Experience integrating ML model inference servers (vLLM, TGI, Triton) into backend services
- Familiarity with audio/voice processing pipelines — WebSocket audio streaming, VAD, codec handling
- Experience with Keycloak or similar self-hosted identity management platforms
- Experience with Weaviate or similar vector databases for RAG systems
- Understanding of healthcare compliance audit requirements and PHI handling
- Experience with Prometheus/Grafana monitoring and ELK stack logging
