Head of Core AI at Physical Superintelligence
Boston, MA
—
Jul 30, 2026
Boston, MA
Sep 24, 2026
What this job asks for AI summary
A leadership role at an early-stage AI-for-physics startup, responsible for building and directing the team that develops the models, agents, and training systems powering the company's core scientific discovery capability. The position is explicitly a player-coach role requiring hands-on technical contributions alongside people management, roadmap ownership, and cross-functional partnership. It suits a research leader with a frontier ML background who can both set strategic direction and personally build and evaluate AI systems.
Principal level · Boston-Cambridge-Newton, MA-NH · Full-time
Quick apply — this platform usually takes a CV and a few fields.
We read this from the posting text with AI. Skim the description below before ruling yourself out.
How this req sits in the market our data
Roughly 60 people in the Boston-Cambridge-Newton, MA-NH area plausibly meet what this posting asks for (computer and information systems managers). range 25–90
Applicant volume Moderate — A normal amount of company. The rare requirements below are what will separate a shortlisted application from the rest.
Rare in this occupation — lead with these, and say what you built with them.
What the occupation pays Median $219,687 (middle half $178,102–$275,447).
Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Aug 1, 2026. It is a model, not a headcount.
Why we read it this way (6)
The role is titled 'Head of Core AI' with no explicit seniority level in the title; the title states no level accordingly. However, the scope — owning an entire research organization, setting company-wide AI roadmap, and reporting at the executive level — clearly supports a Principal-level classification.
SOC classification is a genuine judgment call. The role has substantial people-management and organizational ownership (pointing to 11-3021), but the posting explicitly requires the person to remain hands-on technically, personally build and train systems, and review code — a meaningful IC component. 15-2051 (Data Scientists) is listed as the runner-up given the deep ML/AI research requirements.
Compensation is described only as 'competitive' with salary, benefits, and equity — no figures are given.
The four required skills (RL, agentic systems, large-scale training, evaluation methodology) are drawn from the 'What We're Looking For' requirements section. 'Agentic coding tools' and 'simulators' appear under 'Nice to Have' and 'How We Work' respectively and are marked preferred.
Publications at NeurIPS, ICML, or ICLR are listed as a required signal of caliber but name no specific technology or tool, so they are omitted from the skills list per the generic-concepts rule.
Ignored 2 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): large-scale ML training, ML evaluation methodology.
Read the full posting
The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.