Engineering
Lead AI Engineer — Agents & RAG
Own the design and delivery of grounded agents over private knowledge: retrieval, citations, tool use, approvals, and evals.
Location
India / Global Remote
Workplace
Remote
Type
Full-time
Team
Engineering
Experience
4+ years
Compensation
Compensation discussed during process
Delivery ownership
Build agents that know when evidence is missing.
This role owns the core systems behind Ydhya's agent and RAG engagements: the retrieval layer, orchestration logic, tool contracts, and evaluation harness.
You will work directly with client workflows where incorrect answers, missing citations, or unsafe tool calls can break trust.
What you will do
01
Design and ship RAG pipelines with chunking, metadata, hybrid retrieval, reranking, and citations
02
Build LangChain and LangGraph agent flows that use tools, memory, approvals, and fallback paths safely
03
Create evaluation sets for groundedness, retrieval quality, latency, and workflow completion
04
Deploy on client cloud environments and communicate technical tradeoffs clearly
What we are looking for
01
4+ years building production software or AI/ML systems
02
Strong Python, API design, data pipelines, and cloud deployment fundamentals
03
Hands-on experience with LangChain, LangGraph, vector databases, and retrieval design
04
Practical judgment around evals, observability, security, and human-in-the-loop controls
Apply
Send a clear note about what you have shipped.
Include the role, your resume, and links to work that show your judgment. Short, concrete notes are better than generic cover letters.