This is an early-stage founding role where youll work across the full AI stack - from prompt and context engineering to multi-agent orchestration, evaluation, and production deployment. Youll shape how AI solves real-world security problems and ship systems end-to-end. You will design, build, and deploy autonomous agents with reasoning and tool-use capabilities, create reliable multi-step agent workflows, and partner closely with security experts to encode domain knowledge into scalable AI systems.

Qualifications :

- Strong hands-on experience building LLM-powered systems in production

- Deep experience with agent-based architectures and multi-agent systems

- Expert-level Python with experience building scalable APIs

- Strong systems thinking with focus on reliability, performance, and cost optimization

- Experience shipping AI features from prototype to production

- Comfortable operating independently in an early-stage environment

Must Have :

- LLMs in production (Claude, GPT-4, Gemini)

- Agent frameworks (LangGraph, LangChain, AutoGen, CrewAI, LlamaIndex)

- Advanced prompt engineering (few-shot, chain-of-thought, ReAct)

- Context engineering (RAG/CAG, chunking, indexing, reranking, retrieval)

- Prompt lifecycle management (versioning, evaluation, tracing)

- LLMOps & observability (Langfuse or similar)

- Vector databases (Pinecone, Weaviate, Qdrant, Chroma, FAISS)

- Function calling and structured LLM outputs

- Knowledge graphs and graph databases

- FastAPI or Flask

- SQL/NoSQL systems, async workflows, message queues

- AI security (prompt injection mitigation, guardrails, output validation)

- Performance optimization and cost control for LLM workloads

- Docker and cloud platforms (AWS, GCP, or Azure)

- Multi-tenant SaaS architecture and data isolation

Nice to Have :

- Ownership of major AI systems end-to-end

- Fine-tuning, RLHF, LoRA, or domain adaptation

- GPU optimization and inference acceleration (vLLM, TensorRT)

- Open-source contributions or technical writing

- Strong testing discipline and production hardening experience

Additional Requirements :

- Ability to move from ambiguity to shipped production systems

- High ownership and accountability

- Starting time : 8 : 30 AM PST