Senior AI Engineer
Harbingergroup
- Pune, India
- Senior Level
- 7+ years
- Posted
Skills
Required
- LLM
- Python
- Pydantic
- FastAPI
- NoSQL store / Vector Database
- software engineering
- backend production development
- asyncio
- data validation
- patterns
- AI Productionization
- streaming mechanics
- Server-Sent Events/WebSockets
- SQL / PostgreSQL
- Azure
Preferred
- asynchronous
- API & System Architecture
- REST APIs
- SQLModel / SQLAlchemy
- LLMs
- multi-agent frameworks
- token streaming
- Data Pipelines
- semantic search
Python, fast API with AI
Responsibilities
- API & System Architecture: Design, build, and optimize scalable, asynchronous REST APIs utilizing FastAPI, Pydantic, and SQLModel / SQLAlchemy to serve high-throughput AI features.
- AI Integration & Engineering: Build production-grade workflows around LLMs and multi-agent frameworks. Architect intelligent systems implementing prompt invocation, agentic logic, and token streaming.
- Data Pipelines & RAG: Develop and maintain semantic search, Vector Database structures (e.g., Pinecone, Milvus, pgvector), and advanced Retrieval-Augmented Generation (RAG) systems.
- Production Guardrails: Implement robust system defenses for AI interactions, including request time-limits, fallbacks, exception handling, cost control boundaries, and rate limiting.
- Collaboration: Partner closely with data scientists, ML architects, and horizontal product teams to seamlessly translate raw AI models into responsive user-facing components.
- Engineering Excellence: Champion clean code practices by maintaining excellent CI/CD pipelines, comprehensive testing habits via pytest, and detailed observability/telemetry (logging, tracing LLM calls).
Requirements
- Experience: 7+ years of active software engineering experience with a heavy emphasis on backend production development.
- Python Mastery: Expert-level proficiency in Python, deeply understanding asyncio, data validation via Pydantic, and advanced framework patterns.
- FastAPI Expertise: Proven track record of shipping production-ready FastAPI backends that natively leverage asynchronous workers to manage slow upstream LLM or retrieval calls.
- AI Productionization: Practical experience taking LLM features into real-world production enviornments. Understanding of streaming mechanics (Server-Sent Events/WebSockets), cost optimization, and token limitations.
- Database Design: High proficiency in relational data modeling (SQL/PostgreSQL) combined with experience in at least one NoSQL store or Vector Database.
- Cloud Infrastructure: Hands-on experience with cloud platforms (Azure preferred),
