Senior AI Engineer - Multi Agent Orchestration

Your responsibilities:

Architect and implement multi-agent orchestration patterns using LangGraph,designing skills-based workflows that power the Refinement, Decision, and Coding Agents across the Agent Chain platform.

Design and evolve prompting architectures and RAG strategies that achieve high-fidelity, reliable AI outputs grounded in application-specific knowledge bases via Model Context Protocol (MCP).

Conduct technical design reviews and establish engineering best practices for AI service development, ensuring compliance with EU AI Act, DORA, and internal DevSecOps standards.

Mentor junior AI engineers in the Hyderabad team on prompt engineering techniques, agentic design patterns, and production-quality Python development.

Collaborate closely with architects and product owners to translate architectural concepts into functional, scalable AI services integrated with GitHub EU Cloud and Jira workflows.

Evaluate and benchmark foundation models (Claude, GPT, Gemini) for specific agent tasks, optimising for accuracy, latency, and cost across the agent pipeline.

Your profile:

Bachelor's or Master's degree in Computer Science, AI, or a related field, with 5+ years of experience building AI/ML systems and at least 2 years working with LLM- based applications in production.

Expert-level Python programming skills with hands-on experience in LangGraph, LangChain, and modern AI orchestration frameworks for building multi-agent systems.

Deep expertise in prompt engineering, RAG architectures, and agentic workflow design, with a proven track record of delivering high-quality, reliable AI outputs at scale.

Strong working knowledge of foundation models (Anthropic Claude, OpenAI GPT, Google Gemini) including fine-tuning strategies, context window management, and model evaluation methodologies.

Demonstrated ability to lead technical discussions, and drive engineering standards within a distributed, cross-timezone team.

Nice-to-have: Experience with Model Context Protocol (MCP), DevSecOps toolchains (GitHub Actions, Terraform, Kubernetes), regulated financial environments, or German language skills.