Agent Developer

Experis; Chennai District; Hybrid; INR 400,000 - 700,000; Full time

Job summary

Experis is seeking an Agent Developer in Chennai with 6+ years of experience to design and build autonomous AI agents, prompts, and reasoning pipelines. You will integrate RAG, grounding layers, and evaluation frameworks while collaborating with backend and platform teams to operationalize capabilities.

The role requires strong Python skills, hands-on experience with LLMs and agent orchestration, and familiarity with vector databases for semantic search.

Qualifications

• Bachelor’s or Master’s degree in CS, AI, or a related discipline.
• 6+ years of total experience and 2+ years of Agent development experience.
• Strong Python programming expertise.
• Hands-on experience with LLMs, agent-orchestration frameworks, prompt engineering, and vector databases.
• Experience with RAG patterns, context engineering, and knowledge retrieval.
• Experience developing autonomous, tool-using, planning, and multi-agent systems.
• Familiarity with evaluation frameworks, observability tooling, and AI guardrails.

Responsibilities

• Develop AI agents, workflows, skills, prompts, and reasoning pipelines using agent-orchestration frameworks.
• Build agents for scoping, domain decomposition, planning, test generation, and specification synthesis.
• Integrate Retrieval-Augmented Generation (RAG) and context-retrieval layers to ground agent outputs.
• Design and implement guardrails based on approved technology standards and knowledge base pre-checks.
• Create reusable prompt libraries, agent skills, and evaluation datasets.
• Continuously evaluate model performance, grounding quality, hallucination rates, and reasoning accuracy.
• Collaborate with backend and platform teams to operationalize agent capabilities.

Skills

Python programming

LLM experience

Agent development

Prompt engineering

Vector databases

Evaluation/observability

Education

Bachelor’s or Master’s in CS/AI

Tools

LangGraph/LangChain

OpenAI/Anthropic/Bedrock/Azure OpenAI

Pinecone/FAISS

Embeddings/semantic search

Job description

Position: Agent Developer
Location: Chennai Hybrid Mode

Experience: 6+ Years
Notice Period: Immediate

Job Description

Builds AI agents, prompts, skills, and reasoning pipelines that automate scoping, decomposition, planning, testing, and specification synthesis.

Responsibilities

• Develop AI agents, workflows, skills, prompts, and reasoning pipelines using agent-orchestration frameworks.
• Build agents for scoping, domain decomposition, planning, test generation, and specification synthesis.
• Integrate Retrieval-Augmented Generation (RAG) and context-retrieval layers to ground agent outputs.
• Design and implement guardrails based on approved technology standards, naming conventions, and pre-checks sourced from the knowledge base.
• Create reusable prompt libraries, agent skills, and evaluation datasets.
• Continuously evaluate model performance, grounding quality, hallucination rates, and reasoning accuracy.
• Collaborate with backend and platform teams to operationalize agent capabilities.

Qualifications

• Bachelor's or Master's degree in Computer Science, AI, or a related discipline.
• 6+ years of total experience and 2+ years of Agent development experience.
• Strong Python programming expertise.
• Hands-on experience with LLMs, agent-orchestration frameworks, prompt engineering, and vector databases.
• Experience with RAG patterns, context engineering, and knowledge retrieval.
• Experience developing autonomous, tool-using, planning, and multi-agent systems.
• Familiarity with evaluation frameworks, observability tooling, and AI guardrails.
• Understanding of enterprise integration patterns and cloud platforms.

Mandatory Tech Stack

• Language: Python
• Agent Frameworks: LangGraph / LangChain or equivalent agent-orchestration frameworks
• LLM APIs: OpenAI / Anthropic / Bedrock / Azure OpenAI or equivalent foundation-model APIs
• Agentic RAG & Retrieval: Vector databases such as Pinecone / FAISS, embeddings, and semantic search
• Prompting: Prompt engineering, context engineering, and structured outputs