Job Description
<h3>📋 Description</h3> •Weave accelerates the way therapeutic knowledge is created, organized, and shared.
•The Research team fuels the engine running that mission—turning novel and state-of-the-art AI into practical and impactful capabilities and open science contributions. Your work will:
•Super-charge Weave’s people with internal tools that cut busy-work for product, engineering, and commercial teams.
•Elevate our platform by inventing and benchmarking algorithms that make LLM-powered workflows much more reliable and robust.
•Advance the field through peer-reviewed papers, open datasets, and collaborations with leading academic & industry labs.
•Invent & prototype new ideas: turn white-board ideas into controlled experiments and live demos.
•Push LLM frontiers: combine large-language models with retrieval, symbolic reasoning, or classic ML; optimize for accuracy, reliability, robustness, latency, and cost.
•Measure ruthlessly: Define crisp hypotheses, select fit-for-purpose metrics (incl. human-in-the-loop), run statistically sound studies, and iterate fast.
•Publish & open-source: Write high-quality papers, release evaluation suites, and present at NeurIPS/ACL/ICML—or local meet-ups if that’s your style.
•Seed the future: Maintain a balanced portfolio: near-term product wins, mid-term reliability projects, and “moon-shot” research that could 10× our differentiation.
•Level-up the team: Lead paper clubs, mentor interns & junior scientists, and model best practices in reproducible science and kind collaboration. <h3>🎯 Requirements</h3> •PhD (or equivalent depth) in ML, Physics, Math, Engineering, computational linguistics, or a closely related field.
•Proven research chops: First-author publications at top venues (NeurIPS, ACL, ICML, EMNLP, etc.).
•LLM mastery: experience fine-tuning, prompt-engineering, and augmenting models with external knowledge/tooling.
•Experimental rigor: fluent in benchmarking, statistical testing, and clear visualization of complex results.
•Production-grade coding: Python + ML stacks (PyTorch/JAX), version control, and comfort reading & contributing to large codebases; bonus for database fluency (Neo4j, Postgres, MongoDB).
•Storytelling skills: you can explain a log-probability trick to executives and translate product pain-points into concrete research questions.
•Bonus points for: open-source maintainer experience, safety/bias evaluation expertise, or a successful track record of bridging academia and industry. <h3>🏖️ Benefits</h3> •🏆 Competitive salary and equity packages.
•🧬 Comprehensive health, dental and vision insurance
•🏝️ Take care of you and yours: generous PTO, parental leave, OneMedical, TalkSpace, Teladoc.
•🚀 Career development opportunities within a company entering a growth phase.
•🌎 This position is based in San Francisco with flexibility to occasionally work remote.