Senior AI/ML Engineer: LLM & Agent Stack

Overview

Design and own core components that help enterprise customers run production agentic AI safely and efficiently, including orchestration, model routing, observability, and policy enforcement.

What you'll do

• Architect scalable graph-based agent orchestration and state management.
• Build multi-agent coordination for production workloads.
• Integrate model adapters, LLM gateway hooks, vector databases, tools, and external APIs.
• Improve tracing, benchmarking, observability, token and cost accounting, latency, throughput, and correctness checks.
• Design moderation hooks, human-in-the-loop checkpoints, audit trails, and policy enforcement.
• Mentor junior engineers and run design reviews.
• Work with strategic customers to prototype agentic solutions and translate them into product features.

What you'll need

• 4–10 years of software engineering experience.
• Substantial experience building distributed systems, infrastructure, or ML platforms.
• Production experience integrating and deploying LLMs, including RAG, retrieval, and embeddings pipelines.
• Hands-on experience with LangGraph, LangChain, or custom agent runtimes.
• Experience designing stateful workflows.
• Experience building observability, cost controls, and policy enforcement for production services.

Nice to have

• Contributions to open-source LLM orchestration tools.
• Familiarity with on-premises or hybrid cloud deployments, data residency, and compliance requirements.
• Background in LLM security, privacy, or model governance.
• Leadership in cross-functional projects and direct customer engagement.

Details

• Location: Bengaluru, Karnataka, India.
• Hybrid work with three office days per week: Tuesday, Wednesday, and Thursday.