Staff AI Research Scientist - Enterprise LLMs & Agents

Candidate; San Francisco (CA); Hybrid; USD 200,000 - 260,000; Full time

Benefits offered by this job

Generous PTO

Health coverage (medical, dental, vis

Parental leave

Fertility and family planning support

Wellness stipend

Learning and development stipend

Company off‑sites and 401k/stock

Job summary

WRITER is expanding its AI research team and seeks an experienced scientist to lead work on large language models and agentic systems. You will own research from hypothesis through model training, evaluation, and production deployment, driving multi‑step reasoning and tool use in enterprise workflows.

You will publish at top venues and mentor teammates while collaborating with engineering and product to ship impactful capabilities across hundreds of thousands of users.

Qualifications

• 7+ years hands‑on ML research with large language models and enterprise deployment.
• Publication record in top ML venues or equivalent independent research work.
• Expertise in post-training methods (SFT, RLHF, RLAIF, DPO, GRPO).
• Strong Python and PyTorch, with ability to build and scale pipelines.
• Experience designing or evaluating agentic systems for multi‑step tasks.
• Ability to collaborate across research, engineering and product teams.
• Values: Connect, Challenge, Own, and a drive for impact on real customers.

Responsibilities

• Lead a high‑impact research agenda on LLMs and agentic systems end‑to‑end.
• Design and run large‑scale post‑training experiments (SFT, RLHF, DPO, etc.).
• Develop benchmarks to evaluate complex enterprise tasks and capabilities.
• Create data synthesis pipelines and training signal generation at scale.
• Shape model architecture and roadmap with research, eng, and product teams.
• Publish and present research at NeurIPS, ICLR, ICML, ACL and others.
• Mentor researchers and engineers to uphold scientific rigor.

Skills

ML research

LLM pre-training

Python

PyTorch

Distributed training

Education

PhD in CS/ML/NLP or equivalent

Job description

WRITER is expanding its AI research team and seeks an experienced scientist to lead work on large language models and agentic systems. You will own research from hypothesis through model training, evaluation, and production deployment, driving multi‑step reasoning and tool use in enterprise workflows.

You will publish at top venues and mentor teammates while collaborating with engineering and product to ship impactful capabilities across hundreds of thousands of users.