AI Research Scientist - Enterprise LLMs & Agentic Systems

WRITER; San Francisco (CA); Hybrid; USD 168,000 - 233,000; Full time

Benefits offered by this job

Generous PTO and holidays

Medical, dental, and vision coverage

401k with company stock options

Annual work-life stipends

Job summary

WRITER is seeking an AI Research Scientist to lead and execute research on large language models, agentic reasoning, and system-level capabilities for enterprise-scale AI deployments. You will drive post-training methods, data synthesis, and evaluation to ship impactful capabilities used by thousands of users daily.

The role is hybrid, based in SF or NYC, reporting to the VP of AI Research. Senior applicants are welcome, with compensation scaled to experience and scope.

Qualifications

• 4+ years hands-on ML research, trained models at scale.
• Ph.D. in Computer Science, ML, NLP, or related field, or equivalent demonstrated research with publications.
• Expert-level post-training methods (SFT, RLHF, RLAIF, DPO, GRPO).
• Strong Python and PyTorch (or JAX) with ability to build training pipelines and evaluation infra.
• One or more publications at NeurIPS/ICML/ACL/EMNLP or equivalent.
• Hands-on experience designing or evaluating agentic systems with multi-step reasoning and tool use.

Responsibilities

• Work on high-impact research on LLMs and agentic systems for enterprise AI.
• Design and run post-training experiments (SFT, RLHF, RLAIF, DPO) to improve reasoning and tool use.
• Develop evaluation benchmarks and methodologies for complex enterprise tasks.
• Create scalable data synthesis and curation pipelines for high-quality training signals.
• Shape model architecture and training roadmap with research engineering and product teams.
• Publish and present research at top venues and communities.
• Collaborate with researchers and engineers to maintain high scientific rigor.

Skills

ML research

Post-training methods

Publications

Agentic systems design

Collaborative research

Education

PhD in CS/ML/NLP or equivalent

Tools

Python

PyTorch

Distributed training

Job description

WRITER is seeking an AI Research Scientist to lead and execute research on large language models, agentic reasoning, and system-level capabilities for enterprise-scale AI deployments. You will drive post-training methods, data synthesis, and evaluation to ship impactful capabilities used by thousands of users daily.

The role is hybrid, based in SF or NYC, reporting to the VP of AI Research. Senior applicants are welcome, with compensation scaled to experience and scope.