Prodigy Health
Website:
prodigyhealth.com
Company:
https://www.linkedin.com/company/prodigy-health
Industries: Hospitals and Health Care
Job details:
Job Location: Onsite, Delhi NCR
Experience: 5–8 years (backend), with hands-on AI/LLM production work
Job Type: Full-time
Compensation: As per industry standards (upto 30 LPA)
About the Role
We're looking for a Senior Backend Engineer who can architect and scale a multi-tenant SaaS platform and build production-grade AI systems on top of it. This is a backend-first role for someone who thinks in terms of data models, service boundaries, tenancy isolation, throughput, and cost — not just "calling an API."
You'll own critical parts of the platform: the data layer across MongoDB, PostgreSQL, and Redis; AWS infrastructure and deployment; and the AI systems (RAG, LLM orchestration, extraction, evals) that make our platform genuinely intelligent. You'll set patterns the rest of the engineering team builds on.
What You'll Do
Architecture & scale
- Design and evolve the backend architecture for a multi-tenant SaaS platform serving multiple hospital groups and dozens of units on shared infrastructure.
- Own tenancy strategy: data isolation, per-tenant configuration, noisy-neighbor protection, and scaling patterns that hold up as we add hospitals.
- Design service boundaries, APIs, and async/event-driven workflows for reliability and throughput.
Data layer
- Model and optimize across MongoDB, PostgreSQL, and Redis — choosing the right store per workload (operational data, relational analytics, caching, queues, rate limiting).
- Own schema design, indexing, query performance, migrations, and data integrity across tenants.
AI systems (production-grade, not demos)
- Build and own AI features end-to-end: RAG pipelines (ingestion → chunking → embeddings → retrieval → reranking → citations), extraction, classification, summarization, and copilots.
- Implement LLM orchestration: tool/function calling, structured outputs, deterministic fallbacks, and agent loops only where they earn their keep.
- Build evaluation and guardrails: golden datasets, regression tests, offline + online evals, PII redaction, prompt hardening, and safety rules — critical in a healthcare context.
- Monitor and improve quality, latency, and cost weekly.
Infrastructure & operations
- Own AWS deployment, CI/CD, observability, and cost monitoring for both platform and LLM usage.
- Set up logging, tracing, and alerting so failures and drift are caught early.
Team leverage
- Set reusable backend and AI patterns (prompt/tooling libraries, eval harnesses, service templates) the team can build on.
- Translate business problems into system designs and communicate tradeoffs clearly — accuracy vs cost vs latency, build vs buy, agent vs deterministic workflow.
Must Haves
- 5–8 years of backend engineering with strong fundamentals: clean architecture, API design, testing, and performance tuning.
- Deep expertise in Node.js / TypeScript.
- Strong hands-on experience with MongoDB, PostgreSQL, and Redis in production.
- Proven experience designing and scaling multi-tenant SaaS or enterprise applications.
- Solid AWS experience (compute, networking, storage, deployment) and CI/CD.
- Shipped at least 1–2 AI/LLM features to production, with real understanding of RAG, embeddings, retrieval strategies, structured outputs, and evals.
- Comfort with ambiguity, fast execution, and a bias toward simplicity.
Nice to Haves
- Experience with vector databases (Pinecone, Weaviate, Milvus, FAISS) or Elastic hybrid search.
- Orchestration frameworks (LangGraph, LangChain, LlamaIndex) or equivalent custom systems.
- Docker / Kubernetes and infrastructure-as-code.
- ML fundamentals (embeddings, similarity search, classification, fine-tuning concepts).
- Experience with healthcare data, compliance, or other regulated/enterprise domains.
- Experience with LLM cost monitoring and optimization at scale.
First 90 Days = Success
- 30 days: Ship your first AI feature into CarePro production, with evals and regression tests so quality doesn't drift.
- 60 days: Own a core part of the data or tenancy layer; land a measurable improvement in performance, cost, or reliability.
- 90 days: Leave behind a scalable pattern (backend or AI) the team reuses.
Tech Stack
- Platform: MERN (MongoDB, Express, React, Node.js)
- Data: MongoDB, PostgreSQL, Redis
- Cloud: AWS
- AI: OpenAI / Anthropic models, embeddings + vector DB, orchestration frameworks, evaluation harness
- Dev workflow: Cursor, Claude Code, CodeRabbit, and other AI-assisted coding/review tools
About Prodigy Health
Prodigy Health is an AI-native healthcare technology company building the data and intelligence layer for modern hospitals. We build two products: CarePro, a patient engagement and conversion platform, and Intellica, a health BI cloud for multi-unit hospital analytics. Our platforms are live with leading enterprise hospital groups across India, running mission-critical operations at scale. We're a small, fast-moving team that ships real product into real hospitals — where reliability, security, and data integrity genuinely matter.
We serve multi-unit hospital groups running dozens of facilities on a single platform. That means everything we build has to be multi-tenant, secure, and scalable by default. The core platform is already live on MERN and in production with enterprise hospital chains. We're now scaling the architecture and deepening AI across the product.
If you want to own backend systems end-to-end — architecture, data, infrastructure, and AI — and see them run in production at enterprise scale, let's talk.
Click on Apply to know more.