AI Engineer

Keka Inc.; Ahmedabad District; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Keka Inc. is seeking a skilled AI Engineer to design and deploy production-ready AI agents and multi-agent workflows that automate complex business processes. You will bridge LLM engineering with backend development, using a modern stack centered on LangGraph, PydanticAI, and MCP.

The role focuses on building scalable AI services, integrating with internal APIs, and enabling RAG pipelines across vector databases. Hands-on Python work and MLOps practices are essential in production environments.

Qualifications

• 3–6 years of experience in software engineering, ML engineering, or AI engineering.
• Hands-on experience building production applications in Python.
• Experience with LangChain or LangGraph (or similar LLM frameworks).
• Strong experience with Pydantic and structured data validation.
• Exposure to multi-agent frameworks such as CrewAI is a plus.
• Experience working with LLM APIs and prompt engineering.
• Familiarity with RAG pipelines and vector databases.

Responsibilities

• Design and implement stateful AI workflows using LangGraph.
• Build role-based multi-agent collaborations using CrewAI.
• Develop reliable long-running and branching AI processes.
• Use Pydantic and PydanticAI to enforce type safety and structured outputs.
• Implement schema-driven AI pipelines and validation layers.
• Contribute to reliability, logging, and observability of AI services.
• Build and maintain Retrieval-Augmented Generation (RAG) pipelines.
• Work with vector databases such as Qdrant, ChromaDB, or PgVector.
• Contribute to GraphRAG implementations using Neo4j or FalkorDB.
• Improve search quality using hybrid search and reranking techniques.

Skills

Python

LLM frameworks

LangGraph

LangChain

Prompt engineering

RAG pipelines

Vector databases

Docker

REST gRPC

Cloud functions

Tools

Qdrant

ChromaDB

PgVector

Neo4j

FalkorDB

Ollama

LM Studio

AWS Lambda

GCP Cloud Functions

Job description

We are looking for a highly skilled AI Engineer to help build and scale our internal AI ecosystem. You will design and deploy production-ready AI agents and multi-agent workflows that automate complex business processes.

This role bridges LLM engineering and backend software development, using a modern stack centered on LangGraph, PydanticAI, and Model Context Protocol (MCP).

This is a hands-on role ideal for someone excited about building real-world AI systems and shipping them to production.

Key Responsibilities

Agentic Workflow Development

• Design and implement stateful AI workflows using LangGraph.
• Build role-based multi-agent collaborations using CrewAI.
• Develop reliable long-running and branching AI processes.

Structured AI Services

• Use Pydantic and PydanticAI to enforce type safety and structured outputs.
• Implement schema-driven AI pipelines and validation layers.
• Contribute to reliability, logging, and observability of AI services.
• Build and maintain Retrieval-Augmented Generation (RAG) pipelines.
• Work with vector databases such as Qdrant, ChromaDB, or PgVector.
• Contribute to GraphRAG implementations using Neo4j or FalkorDB.
• Improve search quality using hybrid search and reranking techniques.

MCP & Internal Tooling

• Help build and maintain Model Context Protocol (MCP) servers.
• Integrate AI agents with internal APIs, databases, and tools.
• Support development of internal AI frameworks and reusable components.
• Work with both local models (Ollama / LM Studio) and cloud LLM providers.
• Assist in model evaluation, optimization, and experimentation.
• Support domain-specific fine-tuning and benchmarking.

Performance & Scalability

• Implement semantic caching and context optimization strategies.
• Improve latency, cost efficiency, and scalability of AI services.
• Deploy AI workloads on AWS Lambda or GCP Cloud Functions.
• Write clean, maintainable, production-quality Python code.

Required Skills & Experience: -

• 3–6 years of experience in software engineering, ML engineering, or AI engineering.
• Hands-on experience building production applications in Python.
• Experience with LangChain or LangGraph (or similar LLM frameworks).
• Strong experience with Pydantic and structured data validation.
• Exposure to multi-agent frameworks such as CrewAI is a plus.
• Experience working with LLM APIs and prompt engineering.
• Familiarity with RAG pipelines and vector databases.

Databases

• Experience with at least one Vector DB (Qdrant / PgVector) or Graph DB (Neo4j or FalkorDB).
• Experience with Docker and cloud/serverless deployments.
• Understanding of REST or gRPC APIs.

Preferred Qualifications

• Experience with Human-in-the-Loop workflows.
• Background in semantic search or information retrieval.
• Experience building internal tools or developer platforms.
• Familiarity with model fine-tuning or evaluation.