Full-time · Remote (4+ hours overlap with CET working hours preferred)

Location: Remote (4+ hours overlap with CET working hours preferred).

Type: Full-time.

Application deadline: 3 November 2026, end of day (Anywhere on Earth).

Compensation: $120K – $180K. Adjusted depending on location, seniority, and experience. It can be higher for exceptional candidates or specific locations with higher costs of living (e.g. San Francisco).

Start: As soon as possible.

Set the research direction of one of AI safety's largest on-ramps, and turn it into a pipeline for the field's strongest people.

About Apart

Apart Research is an AI safety nonprofit that builds one of the main global talent pipelines for the AI safety ecosystem. Our remote Research Sprints and Fellowships have brought 8,000+ participants through 50+ Sprints and produced 40+ peer-reviewed papers, including two ICLR 2025 oral spotlights. Alumni work at Anthropic, Google DeepMind, the UK AI Security Institute and other safety labs and government bodies.

Core Responsibilities

Write the research tracks. For every Sprint and Fellowship cohort, produce a miniature open-problems paper: a direction that matters now, three to five problems a team of three can make real progress on in 48 hours (including a toy problem with a known baseline), what strong and weak submissions look like, and what each problem reveals about who attempts it. You convene outside experts to sharpen it and you hold the pen. When a Sprint runs with a research partner, for example an AI control Sprint with Redwood Research, you shape the projects with their researchers as a peer.

Hold the map. Maintain Apart's view of who is working on what across AI safety, which agendas are gaining and losing ground, and where our throughput should and should not go. You read the papers, talk to the people doing the work, and run small checks yourself.

Design every program for talent. Choose the partners, sponsors, universities and local AI safety groups that bring the right people, evaluate who came, and follow up with the best of them.

Decide what runs and prove it works. Own the Sprint calendar and the program portfolio, including new formats such as grand challenges and mini-incubators. Every program gets a theory of change, a premortem and a postmortem. You own evaluation and impact reporting, commission independent review for causal claims, and cut what the evidence does not support.

Lead the team and the numbers. Manage the Special Projects team (Grand Challenges and other new initiatives), the Fellowship Lead, the Sprints Coordinator and everyone else working on the Field Building Division. Propose the contractors and expert collaborations your programs need, and report the Field Building division's KPIs and OKRs to the CEO.

Required Qualifications

You will likely have:

• A real map of AI safety research, with views on which directions matter that you will defend.
• Written research problems for other people: mentor project lists, RFPs, challenge or competition tasks, capstones, etc. Ideally you have a sense of what probelms are tractable in two days vs two weeks.
• Spotted and developed talent: designed a program to reach specific people, evaluated them and followed up, and you know where they are now.
• Started something new, and managed people who delivered it.
• Changed course on evidence, including stopping something you built.
• Enough research ability to judge a paper, run a small experiment and be a credible peer to a frontier-lab research lead, a funder and a senior participant.

Strong networks in AI safety are a strong plus. People who fit well include research program managers at frontier labs, research managers at AI safety organisations, technical grantmakers, challenge and RFP designers, senior field-building organizers and researchers who moved into leading programs.

What We Offer

• A leadership-team seat with saying over the research direction of a whole program portfolio.
• Travel support for conferences and team gatherings.
• A $5,000 annual learning and development budget and protected time for reading, thinking and upgrading your AI workflows.
• A small, fast-moving team that ships and disagrees openly.

Application Process

1. Application form
2. Screening call
3. Take-home assignment (paid)
4. Reference checks
5. Interviews
6. Work trial (paid)
7. Offer

Apart Research is an equal opportunity organization and welcomes applicants of all backgrounds. We may use AI-assisted tools to support review; all applications are read by a human and final decisions are made by people. We process the information you submit to assess your application and may match it against our candidate-sourcing records (public professional information and talent-directory entries) to give reviewers fuller context; we keep application data only as long as needed for recruiting, including considering you for future roles, and delete it on request. Questions about the role or how your data is processed: careers@apartresearch.com.