Lead / Principal GenAI Engineer

This is a forward-deployed AI engineer role sitting within a newly formed, dedicated AI team at a large-cap private equity firm.

You will work directly with our portfolio companies to assess, design, build, and hand off high-impact AI solutions at pace.

The model is deliberately lean. You will work as part of a small, highly specialised pod alongside an AI product manager. Together you are the team. You will be embedded within portfolio companies with full credentials, access, and credibility.

The goal is not to embed indefinitely. It is to get a portfolio company to the point of scaling an AI opportunity and then hand it off, including helping them hire and build their own capability.

• →Assess AI feasibility during diligence, and set technical direction for the systems that follow: data products, retrieval pipelines, model orchestration, agentic workflows, evaluation, deployment, and monitoring.
• →Design and build LLM workflows, retrieval systems, and agentic applications directly against a portco's real enterprise data and infrastructure.
• →Define secure, scalable reference architectures appropriate to each portco's environment, across major cloud and AI platform services as needed.
• →Establish evaluation and observability standards for what you build, test sets, automated evals, tracing, quality scoring, latency/cost monitoring, and regression testing, so handoff teams can trust and maintain it.
• →Build in human-in-the-loop controls and sound data/security practices by default, even where a formal governance function doesn't yet exist at the portco.
• →Develop working prototypes that can evolve into production systems, and transfer that knowledge directly to portfolio company technical teams to support handoff and their own hiring.
• →Evaluate emerging models, frameworks, and tools, and bring back a practical point of view on what's worth adopting versus what's prototype-only.

• 8+ years of professional software development experience with a deep understanding of full-stack architectures.
• Meaningful, hands-on experience shipping production AI/ML systems, not just prototypes, including generative AI or agentic AI.
• Deep fluency in agentic tooling (e.g., Claude Code, Cursor, Codex, GitHub Copilot) and in agentic architecture patterns: single- and multi-agent workflows, supervisor/worker patterns, state and memory management, orchestration, and safe action execution.
• Strong grasp of modern GenAI patterns: advanced RAG, embeddings, vector/hybrid search, re-ranking, tool/function calling, structured outputs, and context engineering.
• Proficiency with modern AI engineering frameworks, LangChain, LangGraph, LlamaIndex, or comparable, and strong Python skills.
• Working knowledge of cloud AI/ML platforms (e.g., Bedrock, Azure AI Foundry, Vertex AI) and enterprise data ecosystems (e.g., Snowflake, Databricks, vector databases) sufficient to work inside whatever stack a given portco already runs.
• Comfort setting LLMOps/MLOps standards on the fly, versioning, CI/CD, containerisation, cost and performance optimisation, in environments that may have none of this today.
• Ability to structure codebases and specifications to produce reliable, context-efficient AI output.
• Sound judgement on architecture tradeoffs, model selection, retrieval strategy, latency, cost, security, vendor lock-in, and the ability to explain those tradeoffs plainly to both technical and non-technical stakeholders.
• Comfortable engaging with both executives and frontline users.
• Forensic attention to detail, extreme ownership, and a positive, can-do mindset.
• Experience on Agile/Scrum teams and on large, complex systems.