U.S. Policy Associate

Lucid Computing•Washington, DC

4d•$120,000 - $195,000•Onsite

About The Position

Lucid is seeking an early-career policy associate to help develop its U.S. policy work, make the state of the art in AI verification legible in Washington, and ensure verification engineers build toward government policy objectives. The associate will work closely with the Head of U.S. Policy and the technical team; the role’s scope will be shaped around the successful candidate. The position connects Lucid’s technical team with policymakers and other Washington stakeholders. It involves explaining accurately what AI verification can do today, what may be possible with investment, what cannot be verified, and the costs of different options. The role also involves tracking domestic frontier AI regulation and communicating government needs to researchers, as well as maintaining records of stakeholder engagement and outcomes. Lucid describes its work as urgent in light of rising frontier AI incidents and developers’ need to demonstrate that they are mitigating extreme risks. One example is retrofitting a 16-GPU H100 cluster in Sweden as an adversarial testbed for verification mechanisms, with repeated cycles of development and independent red-team attacks and open-sourced architecture and results. The role is based in Washington, D.C. Some travel to San Francisco, London, and policy and industry events should be expected. Compensation includes a base salary within a range listed elsewhere in the posting, plus equity; placement in the range depends on experience.

Requirements

Experience in U.S. AI or technology policy in government, Congress, a think tank, or a company policy function. One or two years, including fellowships and internships, is sufficient if the candidate can point to something they helped make happen.
Ability to represent technical work accurately; applicants need not yet be able to read a verification paper or system design unaided, but should be able to do so with some help and recognize when they have made an error.
Writing ability that makes technical material accessible to non-experts.
Comfort operating with limited precedent and a preference for building structure rather than inheriting it.

Nice To Haves

Technical background in security, cryptography, hardware, or distributed systems.
Familiarity with verification regimes in other domains, such as arms control or nuclear safeguards.
Experience making technical or policy work accessible to a wide audience through press, social media, multimedia, and explainers, and a habit of pitching ideas proactively.
Experience in standards bodies or multi-stakeholder governance.
Experience setting up and running systems a small team depends on, such as a CRM for tracking relationships.
Existing relationships across communities working on frontier AI oversight.

Responsibilities

Track federal and state domestic frontier AI regulation and keep the team current on developments relevant to verification.
Draft briefings, public comments, explainers, and responses to requests from policymakers, congressional staff, agencies, and the governance community, accurately describing AI verification capabilities and limitations.
Build and maintain standard materials about Lucid’s work so government and partner requests receive considered answers.
Prepare the team for meetings with policymakers and partners through background research and briefing notes, and follow up by recording agreements and outcomes.
Own the system for tracking engagement and outcomes with external stakeholders.
Work with research and red teams to translate policy proposals into precise verifiable claims and technical results into language usable by policy staff.
Support other needs of a small team, including grant applications and events.