Leads the team this role joins
Heads the group this role sits in
In-house, owns the opening
From your own Gmail and LinkedIn.
About the role
This is a lead or manager-level agentic AI engineering role in Hyderabad, Telangana, India, for engineers who build production AI agents that carry out multi-step tasks from natural language requests. The focus is orchestration, context and harness engineering, MCP tool integration and working with AI coding agents such as Claude Code and OpenAI Codex. Blend is described as an AI services provider working on data science and technology for clients.
What you’ll do
- Design and build LLM-driven agents that plan, choose tools, execute, observe results and retry.
- Build agent workflows with LangGraph, LangChain, Semantic Kernel, AutoGen or similar frameworks.
- Design context engineering strategies and agent harnesses that manage state, tool access, permissions and guardrails.
- Set up repository and project context for AI coding agents, including instructions and automated verification.
- Build MCP servers and tools so agents can take actions in enterprise systems.
- Connect agents to Git, GitHub or GitLab, Artifactory, Slack, databases, APIs and CI/CD platforms.
- Build evaluation pipelines, LLM-as-a-judge methods and regression tests for agent behavior.
- Add human approval steps for high-risk actions and manage token use, latency and cost.
- Deploy and operate agent services built with Python and FastAPI, with monitoring and tracing.
- Turn business requirements into agent solutions with measurable outcomes.
What they’re looking for
- 5-10 years in data science, AI/ML or software engineering, with at least 2 years in generative and agentic AI.
- Hands-on production experience building AI applications and agentic systems.
- Strong Python and API development, including FastAPI or similar frameworks.
- Experience with context engineering and agent harness engineering.
- Experience with AI coding agents such as Claude Code, OpenAI Codex or similar platforms.
- Hands-on experience with A2A and MCP, including connecting agents to enterprise tools and APIs.
- Experience with LLMs, agent orchestration and multi-agent workflows.
- Understanding of agent architecture, function calling, planning, memory and state management.
- Experience with RAG, embeddings, vector databases, chunking, reranking and grounding.
- Working knowledge of Git, CI/CD, Docker, SQL databases and a major cloud platform.
- Understanding of hallucination mitigation, token usage, caching and cost optimization.
What’s on offer
- Competitive salary.
- Career growth with mentorship and hands-on experiences.
- Idea sessions, peer learning chats and recognition programs.
- Company-sponsored certifications in AI, data science, cloud and analytics.
Skills
Questions about this role
Where is this job based?
The listing gives Hyderabad, Telangana, India.
How many years of experience do I need?
The posting asks for 5-10 years in data science, AI/ML or software engineering, with at least 2 years in generative and agentic AI.
Is this a full-time role?
Yes, it is listed as full-time.