Alex Holness-Tofts (Resolution)
This stream will focus on field-building and research amplification, including headhunting and talent projects, ecosystem incentives such as large impact bounties, and operational work to scale an alignment research team.
Stream overview
**Field-Builders** launching headhunting or talent projects. Examples include finding experienced entrepreneurial operators, sourcing people with industry compute expertise, or building an AI safety headhunting company. I'm also interested in mentoring projects on ecosystem incentives, e.g. setting up very large impact bounties which mimic the high-upside structure of startups.
**Amplifiers** to embed into our Research Operations team to scale the impact of our research. You'd work with Resolution founders and Research Program Leads to identify and fix obstacles to us getting ASI aligned. Examples include running an intensive collaboration workshop with top academics, building AI products to automate parts of research or operations, launching or revamping our grantmaking or Fellows program, building a management playbook for researchers new to leading teams, or improving our compute allocation system.
**Founders** building a new AI safety research organisation. I can support with strategy, hiring, co-founder search and execution. I'm not a fit for product or advocacy organisations, and you'd need separate mentorship on research direction.
Mentors
Alex is a co-founder of [Resolution](https://resolution.org/), a research nonprofit using empirics, theory, and automation to get to higher confidence in alignment. He previously worked at the UK AI Security Institute, where he led strategy and operations for the £30m [Alignment Project](https://alignmentproject.aisi.gov.uk/). Before that, he built [80,000 Hours](https://80000hours.org/)’ automated headhunting product, which has made 50+ placements in AI safety and governance. He also co-founded [LASR](https://www.lasrlabs.org/) and [Leaf](https://leaf.courses/), and co-authored the original [Effective Altruism introductory programme curriculum](https://www.effectivealtruism.org/courses/introductory-program).
Mentorship style
Fellows we are looking for
Strong motivation to reduce loss-of-control risks from misaligned superintelligence.
**One** of the following:
1. Impressive experience outside AI safety, for example: successfully scaling a startup as an early hire, 5 years at a top consultancy with above-average progression, or an unusually strong background from a leading tech company.
2. A strong AI safety operational track record, like leading a research fellowship, or talent program.
[Reasoning transparency.](https://coefficientgiving.org/research/reasoning-transparency/)
Project selection
The Winter 2027 cohort offers a wide range of research streams led by experts across AI alignment, interpretability, governance, and safety. Each stream provides its own research agenda, methodology, and mentorship focus.
