Job Description
About the role
You'll build the core agent behaviour and capabilities that power revenue agent: tool-use strategies, context construction, model selection, multi-agent coordination, scoring, and evals. You'll own ambiguous, hard research problems end-to-end: forming hypotheses, designing experiments, building the training/eval/grading infrastructure needed to test them, and pushing results into production.
What You'll Do
- Advance the agent harness - the core loop, tools, environment, guardrails, and model behavior that power the agent. How it selects tools, constructs context, coordinates across
sub-agents, and handles longer-horizon, multi-step tasks. Iterate through systematic experimentation.
- Design and build evaluations for complex revenue tasks (deal health, risk, pipeline questions). Regression detection, benchmarks, automated eval pipelines.
- Build scoring and prediction models: deal health scoring from unstructured data (call transcripts, emails), cross-deal pattern mining, explainable risk signals.
- Research and design reinforcement learning (RL) approaches for improving agent behavior.
- Improve context engineering and data quality - entity resolution, information extraction, knowledge graph and context graph that determine agent reasoning quality.
You may be a fit if you have
- Experience with LLMs or large-scale deep learning systems, including training, fine-tuning, RLHF. You reason about why models behave the way they do, not just how to use them.
- Experience building and evaluating agentic systems - tool use, multi-step reasoning, context management.
- Ability to manage ambiguous research with minimal guidance. You scope the problem, design the experiment, and drive to conclusions.
- Experience designing eval datasets or frameworks, especially for tasks without clean ground truth.
- Publications at top-tier peer-reviewed conferences or journals.
- MS or PhD in CS, math/statistics, or other equivalent quantitative discipline.
What We Offer
- Competitive salary and equity
- Meaningful scope from day one — you'll own things, not just support them
- A sharp, low-ego team that moves fast and communicates directly
- In-office culture in San Francisco — we build together in person
- True founding AE opportunity, you'll define how this company sells
- Tight feedback loop with founders and product
- Clear path to sales leadership as the team scales
- Opportunity to join an AI-first product at a pivotal stage
Additional Benefits
- Comprehensive medical, dental, and vision coverage
- 401(k) plan
- Unlimited PTO
- Daily in-office lunch and fully stocked kitchen
Required Skills
[Research and Analysis]
Additional Information
NA
