How to apply for Local Compute

About Foresight Institute

The Foresight Institute is a San Francisco-based research nonprofit that funds science and technology too early-stage or interdisciplinary for legacy institutions. It is unique in its focus on long-term, high-impact technologies like AI safety and molecular manufacturing, and offers a rare chance to work at the intersection of cutting-edge research and real-world deployment. Working here means contributing to open, public-interest infrastructure that could shape the future of computing and AI safety.

About the role

This grant-funded role is for building accessible, secure local compute infrastructure to democratize AI compute for open science and AI safety research. You will design and prototype local AI compute clusters that non-experts can deploy in diverse connectivity and power environments, and develop software, economic models, and business cases for scaling distribution. The impact is direct: lowering barriers to compute access for researchers who need it most, while aligning with AI safety priorities.

A typical day

A typical day might involve prototyping a low-power compute cluster using single-board computers, testing its deployment in a simulated off-grid environment, and refining software for ease of use by non-experts. You might also spend time developing an economic model for scaling distribution, coordinating with other grant recipients at a Foresight Node meeting, and documenting your work for open-source release.

Who Foresight Institute is looking for

• Proven expertise in hardware, software engineering, or distributed systems, with a portfolio of relevant projects (e.g., single-board computer clusters, mesh networking, or energy-efficient computing).
• Experience designing for non-expert users in challenging environments (e.g., off-grid, low-bandwidth, or intermittent power), with a focus on usability and robustness.
• Demonstrated commitment to open-source principles and AI safety alignment, such as prior contributions to open-source AI safety tools or participation in related communities.
• Ability to develop economic models and business cases for scaling technology distribution, ideally with experience in grant-funded or nonprofit projects.

Tips for this application

• Explicitly state your commitment to open-sourcing all code, data, and outputs, and provide examples of past open-source contributions (e.g., GitHub links).
• Highlight any experience with AI safety alignment, even if indirect (e.g., working on interpretability, robustness, or safe deployment of AI systems).
• Describe specific projects where you designed hardware/software for non-experts in diverse environments, emphasizing constraints like power, connectivity, and cost.
• Demonstrate familiarity with Foresight Institute's mission and past projects; reference their focus on early-stage, interdisciplinary tech and how your work aligns.
• Address the grant nature of the role by outlining a clear, milestone-based plan for what you would achieve with $30k–$100k, including prototype timelines and scaling strategy.

What to cover in your cover letter

['Your technical expertise in building distributed systems or hardware for challenging environments, with concrete examples of prototypes or deployments.', 'Your commitment to AI safety and open science, including any past work that aligns with these values.', 'A specific proposal for how you would design and scale local compute clusters, addressing power, connectivity, and usability for non-experts.', 'Your willingness and ability to actively participate in Foresight Nodes and contribute to the broader Foresight community.']

Research before applying

• Explore the Foresight Institute's website, especially their grant programs, focus areas, and past funded projects to understand their priorities and culture.
• Review Foresight's publications and blog posts on AI safety, compute infrastructure, and emerging technologies to align your proposal with their vision.
• Research the concept of 'Foresight Nodes'—what they are, how they operate, and how you might contribute to them.
• Look into existing open-source local compute projects (e.g., TinyML, federated learning, mesh networks) to identify gaps and opportunities that Foresight might value.

Likely interview topics

Based on the job description, expect questions about:

• Technical design: How would you architect a local AI compute cluster that is deployable by non-experts in areas with unreliable power and connectivity?
• Economic model: What business or economic models could sustain and scale the distribution of these clusters beyond grant funding?
• AI safety alignment: How does your approach to local compute infrastructure contribute to AI safety research and alignment?
• Open-source commitment: How have you handled open-sourcing in past projects, and how would you ensure all outputs are accessible and reusable?
• Community engagement: What ideas do you have for actively participating in Foresight Nodes and fostering collaboration among grant recipients?

Common mistakes to avoid

• Proposing a generic or overly broad project that doesn't specifically address local compute for open science and AI safety; show you understand the niche.
• Neglecting to mention open-source commitment or AI safety alignment, as these are core requirements.
• Ignoring the grant aspect by presenting a project that lacks clear milestones, budget justification, or scalability beyond the initial funding.