Research Engineer/ Applied Scientist

We're looking for exceptional Research Engineers who enjoy turning cutting-edge AI research into production systems used by enterprise customers. This role sits at the intersection of research and engineering - you'll design experiments, invent new approaches, and ship systems that directly impact customer experience.

This role is based out of our San Mateo office.

What You'll Do

• Design and run rigorous experiments, including ablations, benchmarking, and error analysis, to improve NL-to-SQL generation, retrieval, ranking, and agentic reasoning.
• Advance multi-turn conversational understanding through improved context retention, entity resolution, conversational memory, and intelligent clarification strategies.
• Build robust evaluation frameworks, including automated benchmarks, regression suites, and LLM-as-a-judge methodologies to measure and improve model quality.
• Prototype, validate, and productionize ML techniques, including fine-tuning, distillation, retrieval optimization, and agent architectures.
• Evaluate emerging foundation models and AI research, rapidly translating promising advances into production experiments.
• Contribute to the broader AI community through technical reports, or conference talks.

What We're Looking For

• 5+ years of experience in applied machine learning, NLP, large language models, or AI research, with a proven track record of shipping production AI systems.
• Master's degree in Computer Science, Machine Learning, Artificial Intelligence, NLP, or a related field required; PhD is a strong plus.
• Deep experience with modern LLMs and agent architectures.
• Strong software engineering skills with production-quality Python and experience using modern AI frameworks.
• Demonstrated ability to design rigorous evaluation frameworks, benchmark models, and make data-driven decisions balancing quality, latency, cost, and reliability.
• Demonstrated ability to improve model cost/quality through experimentation—including retrieval optimization, fine-tuning, distillation, inference optimization, and agent design.

Nice to Have

• Publications at leading AI or NLP conferences (ACL, EMNLP, NeurIPS, ICML, ICLR, NAACL, CVPR) or meaningful contributions to open-source AI projects.
• Experience adapting foundation models and deploying production-scale LLM systems.

You'll Thrive Here If You...

• Think scientifically and validate ideas through rigorous experimentation.
• Care equally about elegant research and measurable production impact.
• Enjoy digging into production failure cases to uncover new research directions.
• Balance curiosity with pragmatism, you know when to explore and when to ship.
• Want your work to directly influence how enterprises interact with AI every day.

Why WisdomAI

We're building foundational technology at the intersection of large language models, reasoning, and enterprise analytics. Our platform empowers people closest to the business with trustworthy, conversational access to insights hidden across their organization's data.

You'll join a team of engineers and researchers tackling some of the hardest problems in applied AI—from structured reasoning and conversational memory to retrieval, planning, and trustworthy analytical agents.

You'll have the freedom to explore bold ideas, the infrastructure to validate them quickly, and the opportunity to see your work deployed in production for leading enterprises.

If solving unsolved AI problems while shipping real-world impact excites you, we'd love to build with you.

Additional Details:

Compensation: The expected starting base salary for this role is $250,000+, in addition to equity and benefits. Final compensation will be determined based on factors including experience, technical depth, interview performance, and overall alignment with the role. We believe in paying competitively for exceptional talent and leveling compensation based on demonstrated impact.

-----------------------------------------------------------------------------------------------

Our CRAFT

-----------------------------------------------------------------------------------------------