Backend Engineer- AI

Role Overview

We are looking for an AI Engineer who will design, build and deploy production-grade LLM-powered systems. This role combines hands-on LLM workflow development, backend API ownership and end-to-end system delivery in a fast-moving, iterative environment.

Role Responsibilities

LLM Workflow Builder

Design and orchestrate prompts, tools, memory, and agent workflows using frameworks such as LangChain, LangGraph and LangServe.

API & Microservice Owner

Build and maintain high-performance FastAPI services (REST and streaming) that are secure, observable and scalable.

End-to-End Engineer

Own the complete lifecycle of servicesfrom code and containerization to CI/CD, Kubernetes deployment and production monitoringwith a strong bias toward rapid, iterative releases.

Cross-Functional Partner

Collaborate closely with Product and Design teams to scope MVPs, estimate effort and ensure predictable and timely delivery.

Key Responsibilities

Architect, implement, and maintain multi-agent workflows and retrieval-augmented generation (RAG) pipelines.

Write clean, well-typed, and well-tested asynchronous Python code using tools such as pytest and poetry.

Instrument, evaluate, and monitor LLM workflows using tools like LangSmith or LangFuse; build guardrails and regression suites for prompts.

Containerize applications using Docker and deploy via Kubernetes and GitHub Actions, with automated rollbacks and alerting.

Optimize system latency, throughput, and cost; troubleshoot and resolve production issues end-to-end.

Mentor team members on best practices in LLM application engineering and DevOps.

Must-Have Skills

Strong Python expertise with a solid understanding of async I/O.

Hands-on experience using LangChain or LangGraph in real-world projects.

Experience building REST and streaming APIs using FastAPI (or similar frameworks).

Solid DevOps fundamentals, including Docker, basic Kubernetes, CI/CD pipelines, and observability tooling.

Strong Git practices, including pull request hygiene, code reviews, and clear technical documentation.

Nice-to-Have Skills

Experience with lightweight fine-tuning techniques such as LoRA or QLoRA, or embedding model selection.

Familiarity with vector databases such as Postgres/pgvector, Milvus, or LanceDB.

Exposure to serverless or edge deployments (e.g., Fly.io, Cloudflare Workers, AWS Lambda).

Experience with multi-agent planning frameworks or autonomous tool-use research.

Employment Type

• Full-time, Work from Office, Bangalore.