Senior AI Data Engineer

ExlService Holdings, Inc.; Pune District; On-site; INR 4,000,000 - 6,000,000; Full time

Job summary

ExlService Holdings, Inc. in Pune, India, seeks an experienced GenAI engineer to design and deploy LLM-powered applications and agentic patterns for business use cases.

You will build end-to-end RAG pipelines, implement prompt engineering, and deliver production-grade APIs, collaborating with data engineering and MLOps teams to ensure scalable, secure deployments.

Qualifications

• 6-9 years total experience.
• 1-3+ years in hands-on GenAI / LLM application development (production use cases).
• Strong hands-on experience with LLMs (Claude, OpenAI, etc.).
• RAG pipelines and retrieval optimisation.
• GPT + Agentic AI implementation experience.
• LangChain, LangGraph, or similar frameworks.
• Agent orchestration and tool-calling architectures.
• Deep understanding of LLM limitations, evaluation, and optimisation strategies.

Responsibilities

• Design and develop LLM-powered applications using agentic patterns for business use cases.
• Build and optimise end-to-end RAG pipelines (ingestion, embeddings, retrieval, orchestration, response synthesis).
• Implement prompt engineering and orchestration techniques (prompt chaining, tool/function calling, structured outputs).
• Develop production-grade APIs and services (FastAPI/Flask/Streamlit) for GenAI applications.
• Integrate LLM solutions with enterprise systems, data platforms, and workflows.
• Apply guardrails and evaluation frameworks to improve response quality, reduce hallucinations, and ensure responsible AI usage.
• Collaborate with Data Engineering and MLOps teams for data pipelines, deployment, monitoring, and scaling.
• Contribute to reusable components, documentation, and engineering best practices.

Skills

GenAI Development

LLM Applications

Python

Pyspark

API Integration

LangChain

LangGraph

Agent Orchestration

Data Engineering

Cloud Platforms

Tools

LangChain

LangGraph

OpenAI API

Claude

Azure Databricks

Snowflake

FastAPI

Docker

Job description

Key Responsibilities

• Design and develop LLM-powered applications using agentic patterns (single/multi-agent) for business use cases
• Build and optimise end-to-end RAG pipelines (ingestion, embeddings, retrieval, orchestration, response synthesis)
• Implement prompt engineering and orchestration techniques (prompt chaining, tool/function calling, structured outputs)
• Develop production-grade APIs and services (FastAPI/Flask/Streamlit) for GenAI applications
• Integrate LLM solutions with enterprise systems, data platforms, and workflows
• Apply guardrails and evaluation frameworks to improve response quality, reduce hallucinations, and ensure responsible AI usage
• Collaborate with Data Engineering and MLOps teams for data pipelines, deployment, monitoring, and scaling
• Contribute to reusable components, documentation, and engineering best practices

Experience & Core Requirements (Must-Have)

Overall Experience

• 6-9 years total experience
• 1-3+ years in hands-on GenAI / LLM application development (production use cases)

LLM / GenAI & Agentic Engineering

• Strong hands-on experience with:
  • LLMs (Claude, OpenAI, etc.)
  • RAG pipelines and retrieval optimisation
  • GPT + Agentic AI implementation experience
• Experience with:
  • LangChain, LangGraph, or similar frameworks
  • Agent orchestration and tool-calling architectures
• Deep understanding of:
  • LLM limitations, evaluation, and optimisation strategies

Core Engineering

• Strong Python/Pyspark engineering expertise (production-grade development) with proven API integration experience
• Deep data analysis experience and handling large volume of data
• Fabric/Azure Databricks/Snowflake data engineering integration skills
• Good exposure to:
  • Cloud platforms (Azure/AWS/GCP)
  • SQL
  • Containers, CI/CD, monitoring

Data / AI Foundations (Mandatory)

Prior experience in one or more:

• Data Engineering (ETL/ELT, pipelines, orchestration)
• Data Science / ML lifecycle (especially NLP)
• Analytics engineering / data products

Good-to-Have / Preferred

• Experience with fine-tuning techniques (LoRA, PEFT) or prompt tuning strategies
• Experience with enterprise GenAI security & privacy practices (data masking, access control, compliance)
• Familiarity with Azure AI ecosystem (Azure OpenAI, Azure AI Search, Fabric, etc.)
• Exposure to agentic coding tools (e.g., Claude Code or similar environments)