Our AI practice ships systems with audit trails, cost ceilings, and evaluation gates - the unglamorous engineering that separates production AI from conference demos. Youll build RAG pipelines over messy enterprise data, agents that execute real workflows, and the harnesses that prove they work.
You should be the kind of engineer who asks how do we know its right before which model should we use
What you will do
- Design and ship LLM applications: RAG platforms, copilots, document intelligence, agents
- Build evaluation suites that gate releases on accuracy, safety, latency, and cost
- Engineer data boundaries: what leaves the clients environment and what never can
- Integrate AI into enterprise systems through governed, least-privilege interfaces
- Track the model landscape and translate it into client-ready recommendations
What we are looking for
- 3+ years software engineering with strong Python or TypeScript
- Shipped LLM features beyond prototypes: retrieval, orchestration, evaluation
- Working knowledge of embeddings, vector stores, and prompt/context engineering
- Transparent writing - your design docs get read by client architects
Nice to have
- LangGraph or similar agent frameworks in production
- Fine-tuning and model-serving experience (vLLM, GPU infra)
- Classic ML background (forecasting, classification)
Apply Three conversations, and we tell you where you stand after each one.
Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.