Detailed Job Description (Day to Day):*****

AI Agent Development

Hands-on expertise in building agentic applications and multi-agent

workflows using frameworks such as LangChain, LangGraph, Crew AI, OpenAI

Agent SDK, or similar

Strong understanding of agentic patterns: planning, memory, tool usage,

reasoning chains, multi-agent coordination, and failure handling

Proficiency in building **Retrieval-Augmented Generation (RAG)** systems —

embeddings, vector databases, semantic search, chunking strategies

Strong **Prompt Engineering** skills — designing, iterating, and optimising

prompts for reliable agent behaviour

Experience integrating tools, APIs, and data sources via MCP servers,

function calling, and webhook patterns

Ability to implement agent memory, state management, and context window

optimisation

### Evaluation & Quality

Experience designing and implementing **evaluation frameworks** for agentic

systems — measuring reasoning quality, tool usage accuracy, task completion

rates, and failure modes

Ability to build automated test harnesses for agent workflows (unit,

integration, end-to-end)

Experience with LLM evaluation techniques: human-in-the-loop scoring,

automated benchmarks, regression testing for prompt changes

Monitoring and observability for production agents — tracing, logging, cost

tracking, latency profiling