GenAI Engineer

We're looking for a GenAI Engineer who can build production agentic AI systems from scratch. You'll work directly with the founding team to design and implement multi-agent architectures, RAG pipelines, and LLM-powered workflows for enterprise clients. This is a hands-on role. You'll write code daily, not manage teams or sit in meetings.

Responsibilities:

• Build multi-agent systems using LangGraph, CrewAI, AutoGen, or similar frameworks.
• Design and optimise RAG pipelines with hybrid retrieval (dense + sparse + reranking).
• Implement MCP (Model Context Protocol) integrations for tool orchestration.
• Fine-tune LLMs using LoRA/QLoRA/PEFT for domain-specific tasks.
• Optimise inference with vLLM, quantisation, and batching for production latency.
• Build agent memory systems (short-term and long-term) for context persistence.
• Set up LLM evaluation frameworks (RAGAS, custom metrics) for quality assurance.
• Deploy and monitor agentic systems on cloud infrastructure (AWS/Azure/GCP).

Requirements:

• 5-8 years of experience in ML/AI engineering.
• Hands-on experience building agentic AI systems in production (not just POCs).
• Strong proficiency with LangGraph or LangChain for agent orchestration.
• Experience with RAG architectures, chunking, embeddings, vector DBs, and reranking.
• Solid Python skills and familiarity with FastAPI and async patterns.
• Experience deploying ML systems to production.

Strong Plus:

• MCP (Model Context Protocol) implementation experience.
• LLM fine-tuning (LoRA, QLoRA, PEFT, DPO).
• Inference optimisation (vLLM, TensorRT, quantisation).
• Multi-agent orchestration with specialised agents (routers, validators, executors).
• LLM observability tools (Langfuse, LangSmith, Weights and Biases).