Agentic AI Engineer

Agentic AI Engineer

Experience : 4–7 years

Location : Bangalore / New Delhi / Gurugram

Availability : This is an urgent requirement – Immediate joiners preferred

Overview

Design, build, and operate production-grade AI agents and tools using Agentic AI frameworks in Python. You will own agentic workflows end-to-end across planning, reasoning, tool calling, retrieval, evaluation, security, and observability.

Key Responsibilities

Agent Development : Implement agentic workflows in Python using an agent framework (Semantic Kernel preferred; LangGraph, AutoGen, or CrewAI acceptable).

Tool Orchestration : Build reliable tool and function-calling flows with planning, memory, and conversation orchestration.

Retrieval Pipelines : Develop retrieval pipelines for RAG and hybrid search, including ingestion, chunking, embeddings, ranking, query planning, grounding, and caching.

Cloud Deployment : Deploy agents to production on cloud platforms, integrating identity, networking, cost controls, and runtime observability.

Monitoring & Observability : Instrument applications with tracing, metrics, and logs to ensure performance and reliability.

Evaluation Workflows : Establish evaluation workflows using prompt and flow tests for offline, batch, and A / B scenarios.

System Hardening : Collaborate with product, data, and security teams to harden systems using rate limits, retries, timeouts, and circuit breakers.

Documentation & Mentorship : Maintain clear technical documentation and provide mentorship to peers on best practices and debugging.

Must-Have Qualifications

4–5 years of hands-on GenAI application experience before transitioning into agentic system work.

Strong proficiency in Python 3.11+ (typing, asyncio, packaging, testing with pytest, profiling, CI / CD).

Production experience with at least one agent framework ( Semantic Kernel preferred ; LangGraph / AutoGen / CrewAI acceptable).

Proven expertise in tool / function calling , schema design, argument validation, and multi-step planning.

Experience with retrieval systems using vector stores and hybrid search, including grounding and retrieval evaluation.

Cloud deployment experience with containers, secrets / identity management, networking, monitoring, and alerting.

Strong skills in observability and evaluation (tracing, metrics, log aggregation, experiment design, promotion criteria).

Solid understanding of security and safety fundamentals : P rompt-injection defenses, content policy enforcement, sandboxing, and PII handling.

Clear and effective technical communication with strong collaborative and review practices.

Good-to-Have Skills

Multi-agent patterns (task decomposition, coordinator–worker, human-in-the-loop).

Deeper Azure experience, including Azure AI Search and related AI platform services.

Evaluation experience with regression suites, red teaming, and guardrails .

Proficiency in search / data stores (Elasticsearch, Pinecone, pgvector, Azure AI Search).

Frontend integration for agent UIs with streaming / tool traces and secure API design.

DevOps proficiency with Docker, Kubernetes, GitHub Actions / Azure DevOps, IaC, and secrets management.