Senior Data Scientist - AI Agents Team - Surf AI

5+ years of experience as a Data Scientist or Applied ML Engineer with a strong track record of shipping production models. Proficiency in Python and experience with PyTorch , LangChain , and related LLM frameworks. Proven experience with LLM operations -prompting, embeddings, fine-tuning, RAG, and evaluation. Solid understanding of data modeling , experimentation , and performance measurement for ML-driven products. Experience collaborating in cross-functional teams of engineers, researchers, and product managers. Why Join Us? If you want to work on foundational systems, ship AI into production, and help define how agentic security actually operates, this is an opportunity to do it early and with real ownership.

Develop and deploy LLM-based systems, including conversational agents, retrieval-augmented generation (RAG), and recommendation engines. Lead applied research initiatives from ideation to production, focusing on measurable business and product impact. Build robust evaluation pipelines for prompt optimization, model quality, and agent reliability. Collaborate with backend and AI engineers to integrate models into scalable, monitored production environments. Contribute to our agent framework by designing workflows, memory mechanisms, and multi-step reasoning capabilities. Stay current with emerging LLM and agentic techniques, translating research into practical, production-ready solutions.

5+ years of experience as a Data Scientist or Applied ML Engineer with a strong track record of shipping production models. Proficiency in Python and experience with PyTorch , LangChain , and related LLM frameworks. Proven experience with LLM operations -prompting, embeddings, fine-tuning, RAG, and evaluation. Solid understanding of data modeling , experimentation , and performance measurement for ML-driven products. Experience collaborating in cross-functional teams of engineers, researchers, and product managers. Why Join Us? If you want to work on foundational systems, ship AI into production, and help define how agentic security actually operates, this is an opportunity to do it early and with real ownership.