Back to jobs

[Expression of Interest] Founding Lead, Speculative Prototyping (Range Systems)

New York City, NY; San Francisco, CA
Apply

About Anthropic

Anthropic’s mission is to build reliable, interpretable, and steerable AI systems that are beneficial to society. We are a fast-growing group of researchers, engineers, policy practitioners, and operators working together on high-consequence technical and social questions.

About the Team

We are exploring a founding leadership role for a new Speculative Prototyping function. This role is designed for people whose strongest advantage is range: those who can work fluently across engineering, design, research, operations, and public explanation, and who can turn that breadth into practical institutional outcomes.

You will define how speculative artifacts are commissioned, built, tested, and integrated into real decision processes across research, product, policy, and communications. This work is not prediction theater. It is disciplined prototyping under uncertainty: building plausible, inspectable futures early enough to improve choices before they harden.

We are especially interested in candidates who can carry both conceptual and material freight: from framing and pedagogy to implementation, operationalization, and delivery.

What You Would Build and Steward

  • Establish the operating model for Speculative Prototyping, including scope, intake, review, prioritization, and quality criteria.
  • Build a small interdisciplinary team that combines technical prototyping, narrative framing, evaluation design, pedagogy, and decision support.
  • Create repeatable pipelines that convert ambiguous future questions into concrete artifacts, simulations, and testable hypotheses.
  • Translate artifacts into usable guidance for launch decisions, safety posture, policy options, and organizational learning.
  • Design internal pedagogies so partner teams can read, critique, and apply speculative artifacts with rigor.
  • Own delivery from first framing through stakeholder adoption, including communication to mixed technical and non-technical audiences.

Representative Programs

  • Launch a standing “futures-to-tests” program that converts scenario narratives into benchmarkable evaluation suites.
  • Run a cross-functional sprint on an emerging autonomy behavior and deliver artifact-backed options with explicit tradeoffs.
  • Create a governance rehearsal artifact set for executive and policy audiences that includes failure paths and intervention points.
  • Develop an internal teaching module that helps teams distinguish compelling narrative from evidence-bearing prototype.

You Might Be Right for This If Your Work Shows

  • A demonstrated pattern of high-quality work across multiple domains rather than a single narrow lane.
  • Strong engineering fluency (Python and modern AI tooling) and the ability to move from concept to working system.
  • Evidence of founder/operator ownership, or equivalent 0->1 responsibility under ambiguous conditions.
  • Experience turning complex ideas into legible artifacts that influence real decisions.
  • Strong writing and communication in both technical and public-facing registers.
  • Judgment in high-stakes settings where safety, policy, or social impact constraints materially matter.
  • Evidence of repeated reinvention: learning new domains quickly and productively without sacrificing craft.

Especially Valuable Signals

  • Track record of combining technical execution with cultural or creative practice (for example publishing, media, documentary, photography, fabrication, or archival work).
  • Experience designing or facilitating workshops, curricula, or community learning environments for complex technical topics.
  • Comfort moving across product, research, and policy contexts without losing technical rigor.
  • Experience building systems that include human and agentic collaborators, with explicit safeguards and verification loops.
  • Evidence of long-horizon commitment to public-interest or institution-shaping work.

Annual Salary:

The annual compensation range for this role is listed below.

$430,000 - $990,000 USD

Operating Conditions

  • Education requirements: Bachelor’s degree in a related field or equivalent practical experience.
  • Location-based hybrid policy: We currently expect staff to be in an Anthropic office at least 25% of the time; some projects may require more in-person collaboration.
  • Visa sponsorship: We sponsor visas where feasible for role and candidate alignment.
  • Relocation: We are open to relocation support on a case-by-case basis.
  • Interview format: For this role, we conduct all interviews in Python. Interviews run in a private, time-boxed Docker environment using synthetic, role-aligned tasks.

How we're different

This role is designed for builders who are difficult to classify by a single specialty, and effective precisely because of that. We value breadth that compounds, not breadth as ornament.

You should apply if your career arc shows repeated translation across systems, media, communities, and technical stacks, with evidence that your range improves execution quality.

Come work with us!

This is an expression-of-interest posting. We may engage candidates on different timelines as team shape and planning evolve.

Candidates may work with an individual assistant or an agent orchestra (multiple agents they regularly orchestrate). We evaluate policy design, intent clarity, verification practice, and collaboration quality, in addition to technical execution.

Interview tasks are hypothetical and do not use active production projects.

Anthropic is a public benefit corporation headquartered in San Francisco. We offer competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a collaborative office environment.

Guidance on Candidates' AI Usage: Learn about our policy for using AI in our application process.

We encourage you to apply even if you do not meet every listed qualification. Strong candidates come from many routes and disciplinary combinations.

Your safety matters to us. Anthropic recruiters only contact candidates from @anthropic.com addresses or clearly identified partner agencies. We will never request fees or banking information prior to employment.

As set forth in Anthropic’s Equal Employment Opportunity policy, we do not discriminate on the basis of any protected group status under applicable law.

Create a Job Alert

Interested in building your career at Anthropic? Get future opportunities sent straight to your email.

Create alert

Apply for this job

*

indicates a required field

Resume/CV

Accepted file types: pdf, doc, docx, txt, rtf

We invite you to review our AI partnership guidelines for candidates and confirm your understanding by selecting “Yes.”

Voluntary Self-Identification

For government reporting purposes, we ask candidates to respond to the below self-identification survey. Completion of this form is entirely voluntary and will not be considered in the hiring process.

As set forth in Anthropic’s Equal Employment Opportunity policy, we do not discriminate on the basis of any protected group status under any applicable law.

Race & Ethnicity Definitions

If you believe you belong to any of the categories of protected veterans listed under VEVRAA, please indicate by making the appropriate selection.

Voluntary Self-Identification of Disability

Form CC-305
Page 1 of 1
OMB Control Number 1250-0005
Expires 04/30/2026

Why are you being asked to complete this form? We are a federal contractor subject to legal reporting obligations. This information is confidential and not visible to hiring decision makers.

How do you know if you have a disability? A disability is a condition that substantially limits one or more major life activities. Disabilities include, but are not limited to, mobility, sensory, neurodivergence, chronic illness, and mental health conditions.

PUBLIC BURDEN STATEMENT: According to the Paperwork Reduction Act of 1995 no persons are required to respond to a collection of information unless such collection displays a valid OMB control number. This survey should take about 5 minutes to complete.