Jobs at Birlasoft (Now Hiring) — 175 open — Page 8

Area(s) of responsibility

Experience: 8+ years | Team Leadership: 2+ years
Primary Tech: Python, GenAI, Agentic AI, LangChain, LangGraph, Azure

Summary

We are seeking a hands‑on AI/ML Lead to drive the architecture, development, and scaling of our enterprise GenAI Platform, with a strong focus on LLM gateway strategy (hosted vs on‑prem), safety/guardrails, and cost/performance optimization. The ideal candidate has deep experience in Azure‑based AI/ML platforms, has led large engineering teams, and demonstrates strong technical leadership in GenAI, agentic systems, Python, and modern orchestration frameworks such as LangChain/LangGraph. This role is suited for a senior technologist who can align strategy, execution, and delivery across complex, end‑to‑end AI/ML lifecycle projects.

Roles & Responsibilities

• Lead design and implementation of an enterprise LLM gateway, including hosted vs. on‑prem strategy, model routing, fallback, and performance SLAs.
• Architect and deliver agentic AI workflows using LangChain/LangGraph with robust observability, evaluation, and safety controls.
• Own safety and guardrails, including prompt security, PII redaction, content filtering, jailbreak protection, and Responsible AI compliance.
• Implement RAG architectures, vector search, embedding strategy, and grounding with enterprise data.
• Optimize cost and performance: token usage modeling, caching, batching, quantization, GPU utilization, and scaling.
• Oversee end‑to‑end SDLC for AI/ML projects: discovery → PoC → production → monitoring.
• Build and maintain production‑grade systems using Python, FastAPI, CI/CD pipelines, containers, and Azure cloud services.
• Manage and mentor a multi‑disciplinary engineering team; drive hiring, technical reviews, delivery quality, and execution velocity.
• Partner with Product, Architecture, and Security teams to define roadmap, assess risks, and deliver business value.
• Ensure enterprise‑grade security, governance, and compliance across all AI workloads.