Full Stack AI Engineer

Experis; Chennai District, Bengaluru; On-site; INR 150,000 - 250,000; Part time

Job summary

Experis is seeking a Full Stack AI Engineer on a contract-to-hire basis to build autonomous, LLM-powered agents in-house. You will design, implement, and ship agents that reason over internal data, call tools and APIs, and carry multi-step tasks to completion.

The role involves hands-on Python development, integration with Azure/AWS services, and experience with retrieval pipelines, vector stores, and multi-agent orchestration. Immediate joining preferred.

Qualifications

• 34 years of software engineering experience with strong, production-grade Python.
• At least one LLM application built beyond tutorial level, such as a chatbot, RAG system, or agent, with the ability to explain design decisions.
• Working knowledge of LLM APIs such as OpenAI, Anthropic, or open-source models, including function/tool calling and structured outputs.
• Strong SQL and relational database fundamentals, including schema design, joins, and query performance.
• REST API development and integration using FastAPI, Flask, or equivalent.
• Strong Git knowledge and familiarity with Docker.
• Practical understanding of LLM agent failure modes, including hallucinations, context limits, tool-call errors, and runaway loops.

Responsibilities

• Build LLM-powered agents from scratch, including prompt design, tool/function calling, memory, and multi-step task orchestration.
• Connect agents to internal databases and APIs, including SQL and data-access layers.
• Implement retrieval pipelines (RAG), including chunking, embeddings, vector search, and relevance tuning.
• Build evaluation harnesses to measure agent accuracy, latency, cost, and failure modes.
• Implement guardrails, retries, error handling, and tracing to ensure predictable agent behavior in production.
• Deploy and monitor agents on cloud infrastructure in collaboration with platform and data teams.

Skills

Python

LLM integration

API design

RAG pipelines

Tool calling

SQL fundamentals

Git

Docker

Tools

SQL

FastAPI

Docker

Git

LLM APIs

Azure

Job description

Position: Full Stack AI Engineer

Job Type: Contract to Hire(C2H)

Experience: 3+ Years
Location: Bangalore & Chennai
Joining: Immediate Joiners Preferred

Role Overview

We are building autonomous, LLM-powered agents in-house from the ground up, with no off-the-shelf vendor platform. Working alongside senior engineers, you will design, build, and ship agents that reason over internal data, call tools and APIs, and carry multi-step tasks through to completion. This is a hands-on build role where you will write production-grade Python every day and see your agents go live.

Key Responsibilities

• Build LLM-powered agents from scratch, including prompt design, tool/function calling, memory, and multi-step task orchestration.
• Connect agents to internal databases and APIs, including SQL and data-access layers.
• Implement retrieval pipelines (RAG), including chunking, embeddings, vector search, and relevance tuning.
• Build evaluation harnesses to measure agent accuracy, latency, cost, and failure modes.
• Implement guardrails, retries, error handling, and tracing to ensure predictable agent behavior in production.
• Deploy and monitor agents on cloud infrastructure in collaboration with platform and data teams.

Essential Skills

• 34 years of software engineering experience with strong, production-grade Python.
• At least one LLM application built beyond tutorial level, such as a chatbot, RAG system, or agent, with the ability to explain design decisions.
• Working knowledge of LLM APIs such as OpenAI, Anthropic, or open-source models, including function/tool calling and structured outputs.
• Strong SQL and relational database fundamentals, including schema design, joins, and query performance.
• REST API development and integration using FastAPI, Flask, or equivalent.
• Strong Git knowledge and familiarity with Docker.
• Practical understanding of LLM agent failure modes, including hallucinations, context limits, tool-call errors, and runaway loops.

Cloud & Data Stack

Azure

• Azure OpenAI Service
• Azure AI Foundry
• Azure AI Search
• Azure Functions
• Cosmos DB
• Azure SQL
• Azure Container Apps

AWS

• Amazon Bedrock
• Amazon OpenSearch or Kendra
• AWS Lambda
• DynamoDB
• RDS/Aurora with pgvector
• ECS/Fargate

Vector Stores

• pgvector
• Pinecone
• Qdrant
• Weaviate
• Chroma

Desirable Skills

• Agent and orchestration frameworks: LangGraph, LangChain, CrewAI, AutoGen, or Semantic Kernel.
• Multi-agent workflows, Model Context Protocol (MCP), async, or streaming architectures.
• LLM observability and evaluation tools such as LangSmith, Langfuse, Arize, or Ragas.
• CI/CD, Infrastructure as Code using Terraform or Bicep, or MLOps/LLMOps deployment experience.
• TypeScript or Node.js for product-facing agent interfaces.
• Open-source contributions or a public portfolio.