
Harness Lead
San Francisco, CAFull-time$140–220K/yrSeen 5 days agoStill listed 5 days ago
Most applications go out cold — see where you stand first. No sign-up to start.
Watch jobs like this. New roles like this one near San Francisco, CA, by email.
Don't just apply. Show up ready.
Olive works from this exact posting.
At a glance
Olive lists jobs from US employers, including remote roles you can work from the United States.
Job overview
Autumn is building and maintaining a research-agent harness that uses skills, automatic data lineage, and internal tools to help models find and use evidence. The Harness Lead will guide harness design and post-training experiments, using production traces and evaluations to improve quality, speed, and cost. The role also works with the Search Lead to identify whether system issues stem from retrieval or tool use.
Skills & qualifications
Skills
Qualifications
Full job description
This role. We build and maintain our own harness. Today, it delivers quality through skills, automatic data lineage, and a curated SDK of internal tools. The harness needs to help the model find evidence, use it correctly, and search for information strategically. The design of harness features is guided by production traces with a single goal: quality. We think of this infrastructure as scaffolding. Our long-term goal is to post-train recurring harness capabilities into the coding model, then remove scaffolding where evaluations show it is no longer needed. This cycle of adding features to the harness and then removing them after “training” them into the model lets us continually improve the agent. We run this cycle monthly. Your goal is to help us move faster and reduce bloat. You'll help decide what belongs in the harness, when to add tools, what the model should learn, and how we can tell when a change actually improves the research. Concretely:
- Harness architecture and execution. You will own the design for our research agent harness. This involves coming up with data-driven features to improve quality and shipping them into production for users.
- The post-training cycle. You are responsible for the design of post-training experiments to teach the model to search better while retiring unnecessary parts of the harness.
- Benchmarks. You will design evals from production traces to measure our agents’ performance across: (1) quality, (2) speed, (3) cost.
- Co-optimization for search. This involves working closely with the Search Lead to distinguish regressions/gaps from poor retrieval or poor tool use, and improve the system at the layer where it matters.
What we are looking for. At Autumn, we strongly believe that successful people possess two traits: (1) agency, (2) ruthless prioritization. You must be able to will things into existence. This requires an incredible drive to make things happen quickly. You must also know the difference between what is urgent and what is important, and when to prioritize what. For this role specifically, we are looking for the right taste for building a harness and training a model within it. Taste comes from experience; we would love to see your work that demonstrates this. We are also looking for work that demonstrates obsession; have you spent an insane amount of time or effort building or improving something? Urgency. All ambitious ideas come with a clock. We are in a rush to build a generational company. If you don’t feel this rush, this may not be the best fit for you.
Similar jobs, posted recently
Open roles like this one, listed in the last 30 days.
Senior Harness EngineerAtoms · San Francisco, CA · $139–172K/yrPosted 6 days agoPosted 6 days ago
Annotations Portfolio LeadPhysical Intelligence · San Francisco, CAPosted 2w agoPosted 2w agoEngineering Lead, USNamespace · San Francisco, CA (Hybrid) · $200–225K/yrPosted 2w agoPosted 2w ago
Carrier Partnerships LeadHarper · San Francisco, CA · $150–220K/yrPosted 3w agoPosted 3w ago
Lead Strategic Partnerships ManagerIronclad · San Francisco, CA (Hybrid) · $180–225K/yrPosted 3w agoPosted 3w ago
You've read the whole posting — now see how you match it.