Overview
Build and own AI and automation projects end to end, from process discovery with operations teams through design, build, testing and deployment, inside a secure, HIPAA-aligned environment that handles protected health information.
What you'll do
- Own AI projects end to end, from understanding the process with operations teams to design, build, testing and deployment.
- Build Python / FastAPI services with a React front end and a PostgreSQL or Azure SQL data layer.
- Integrate LLM APIs into production: prompt design, structured output, RAG over internal documents, and conversational assistants.
- Build agents that run on NeoForge, and automation workflows in n8n, including migrating the existing Power Automate portfolio.
- Build document and data extraction automations (PDFs, EOBs, bank statements) and validate accuracy against a test set before release.
- Deploy and run what you build on Azure, including containerised services, background jobs and scheduled workers, inside a HIPAA-aligned environment.
- Maintain and extend existing production systems, and document your work so the business can run it without a developer.
What you'll need
- 1–2 years of hands-on experience (minimum 1 year) building and shipping software, including at least one AI or LLM project that reached real users, not only coursework or a tutorial build.
- Strong Python, and experience building APIs with a modern framework (FastAPI preferred).
- Practical experience calling LLM APIs (Claude, OpenAI, Azure OpenAI or similar): prompt design, structured output, and handling cost, latency and failure cases.
- Has built at least one retrieval (RAG) system end to end, including embeddings, chunking and vector search (pgvector or similar).
- Experience building LLM agents with tool / function calling (LangChain, LangGraph or similar).
- Workflow automation with n8n, including LLM calls inside workflows.
- SQL and relational databases (PostgreSQL, Azure SQL or equivalent) and REST API integration with third-party services.
- Docker, and deploying and running a service on a cloud platform (Azure strongly preferred).
- Background and asynchronous job processing (Celery, Redis queues or equivalent).
- Working front-end knowledge (React or similar), enough to build and wire up a usable interface.
- Git, and the habit of writing clean code another developer can pick up.
- Good spoken and written English, with ownership and timeline discipline.
- B.Tech, BCA or MCA (Computer Science), completed. Candidates currently pursuing a degree or with backlogs are not eligible.
Nice to have
- n8n, Power Automate, or similar workflow automation tooling.
- OCR and document AI services (Azure Document Intelligence or similar).
- LLM gateway and observability tooling (LiteLLM, Langfuse or similar).
- PII / PHI de-identification (Presidio or similar) and SSO / authentication integration (Entra ID).
- SharePoint and Microsoft Graph integration.
- Speech-to-text (Deepgram, Whisper) or conversational AI (Bot Framework, Copilot Studio, Teams apps).
- Model Context Protocol (MCP), CI/CD pipelines and code-quality tooling such as SonarQube.
- Exposure to US healthcare, HIPAA, or any regulated-data environment.
- A GitHub profile or portfolio.
Details
- Location: Gurugram, Haryana, India. Onsite, 5 days a week.
- Shift: 1:30 PM to 10:30 PM IST, with some flexibility as required.
- This is a backfill position; immediate joiners or candidates on a short notice period are strongly preferred.
Read the full description and apply on the company’s own careers page.