Senior AI Engineer – LLM, RAG & Agent Systems
About the Role
We are looking for a Senior AI Engineer to own the AI layer of our data platform—building production-grade LLM applications, retrieval systems, intelligent agents, and natural-language interfaces over enterprise data.
You will work across RAG, embeddings, vector and hybrid search, agent/tool-calling architectures, LLM evaluation, and self-hosted open-weight models.
This is a hands-on engineering role for someone who has moved beyond prototypes and has built, deployed, and operated LLM systems in production.
What You’ll Own
- Design and build production-grade LLM applications and RAG systems.
- Own retrieval architecture including chunking, embeddings, vector search, hybrid search, and reranking.
- Build agentic and tool-calling systems with appropriate permissions, scoping, validation, and guardrails.
- Develop natural-language interfaces over enterprise data and structured databases.
- Build and maintain LLM evaluation frameworks, including test sets, regression suites, grounding, hallucination, and answer-quality evaluation.
- Work within our data platform and engineering stack rather than relying solely on hosted AI APIs.
- Deploy and optimize self-hosted open-weight models using technologies such as vLLM or equivalent serving infrastructure.
- Optimize inference performance, GPU utilization, latency, throughput, and cost.
- Explore and implement fine-tuning or model adaptation when appropriate.
- Collaborate with data and software engineers to turn AI capabilities into reliable production products.
Required Qualifications
- 5+ years of software or data engineering experience.
- At least 2 years of hands-on experience building and deploying production LLM-based systems.
- Strong Python engineering skills.
- Deep understanding of RAG and retrieval architecture:
- Chunking strategies
- Embeddings
- Vector databases/search
- Hybrid search
- Reranking
- Retrieval evaluation
- Experience building LLM agents or tool-calling systems.
- Understanding of permissions, access control, scoping, validation, and guardrails for AI systems.
- Strong understanding of LLM evaluation, including test datasets, regression testing, grounding, and hallucination detection.
- Experience working directly with data platforms, databases, or enterprise data, rather than only consuming hosted LLM APIs.
- Strong software engineering fundamentals and experience taking systems from prototype to production.
Strongly Preferred
- Experience with self-hosted open-weight models.
- Production experience with vLLM or equivalent model-serving infrastructure.
- Understanding of GPU resource management and inference optimization.
- Experience with fine-tuning, LoRA, or other model-adaptation techniques.
- Experience with Text-to-SQL systems.
- Experience designing or using a semantic layer over real enterprise data models.
- Experience combining unstructured documents with structured enterprise data in a single AI application.
What Success Looks Like
You will be successful in this role if you can build an AI layer that is:
- Accurate — answers are grounded in enterprise data.
- Reliable — quality is measured through automated evaluation and regression testing.
- Secure — agents and tools respect user permissions and data boundaries.
- Scalable — models and retrieval infrastructure perform reliably in production.
- Maintainable — AI capabilities are built as production software, not isolated experiments.
- Useful — users can interact naturally with complex enterprise data.
