Proven strong experience with JavaScript, TypeScript, and Python Advanced GenAI/LLM expertise: RAG, embeddings, vector search, prompt engineering, evals, guardrails, caching, chaining, function calling, etc Fine-tuning expertise: LoRA/QLoRA, instruction tuning, RLHF, domain adaptation, and custom model training Agentic architecture design: multi-agent systems, tool orchestration, planning/reasoning frameworks, memory systems, and agent coordination patterns Eval systems: automated evals, human feedback loops, benchmarking, A/B testing, safety testing, hallucination detection, and performance metrics Experience with LLM frameworks (LangChain, LlamaIndex, LiteLLM, CrewAI, Mastra, etc) and API integrations (OpenAI, Anthropic, etc.) CI/CD pipelines and IaC tools (Terraform, CloudFormation, CDK) Cloud platforms (AWS, Azure, GCP) for GenAI workloads and API management Containerisation and orchestration (Docker, Kubernetes) for AI applications LLM hosting and inference optimisation (vLLM, TGI, local deployment) Vector databases (Pinecone, Weaviate, Chroma, Qdrant) and traditional databases (PostgreSQL, Neo4j) Data pipeline design for RAG and knowledge ingestion Real-time data streaming and processing Document processing and embedding generation workflows React or modern UI frameworks API design for LLM services and conversational interfaces Multiple protocols (HTTP, GraphQL, gRPC, WebSockets, MCP) Real-time chat and streaming response handling Building high-performance, scalable AI applications LLM cost optimisation, token management, and caching strategies Performance monitoring, observability, and debugging production AI systems Comprehensive testing: unit, integration, AI behaviour, security, accessibility