Project summary

ATLAS is a searchable directory of everyone in AI safety, the people and the orgs, in one place you can actually query.

Right now it holds 754 real people and 183 real orgs. Each person has a profile, a photo, a focus area, and links. You can search and filter the whole thing instantly, by field, focus area, org.

This grant makes it comprehensive, fresh, and queryable: a weekly scraper that keeps real people and orgs current from public sources (LinkedIn, personal sites, GitHub, arXiv, the Alignment Forum), AI that drafts each profile and tags its focus area, semantic search, claim-your-profile editing, and an API.

The idea comes from Austin Chen's "Sixteen schemes for AI safety", #1 (Triplebyte for AI safety jobs) and #2 (a database of every AI safety person). Money is pouring into the field, so money stops being the bottleneck and finding the right people becomes it.

What are this project's goals? How will you achieve them?

The field is impossible to see. Recruiters can't find talent, newcomers can't find roles or mentors, people can't find collaborators, independent researchers can't find funders, and nobody can see the shape of the field, where the work is and where the gaps are. People have tried databases and org-maps, but databases rot because nobody updates a spreadsheet, and the maps show orgs, not people, and don't let you query anything.

The goal: kill the search cost for all of that at once, over real and comprehensive data, and put the global south on the map instead of treating it as an afterthought. The two questions a funder actually asks, "who do I reach out to for this role" and "who should I invite to speak on this topic", should take one search.

What "done well" means, and what I'm building toward:

• Comprehensive: every relevant person and org is in it. Started with the full Manifund roster; the scraper widens it from there.
• Accurate and fresh: a weekly refresh keeps it current, with provenance badges showing where each profile came from and whether it's been claimed.
• Fast: a lightweight table, virtualized lists, and a CDN.
• Useful: direct and semantic search, filters on every field, real Projects and Jobs tabs, and a queryable API.
• Trusted: opt-out and removal from day one, claim-your-profile editing behind magic-link auth, and moderation.

Three-month plan:

Month 1: fill the data gaps so coverage is real, ship the Projects and Jobs tabs, auto-tag everyone's focus area, and turn on the weekly refresh.

Month 2: real AI matching and semantic search, claim-and-edit profiles with magic-link auth, moderation, and provenance badges.

Month 3: richer profiles, a queryable API, speed and scale (virtualized lists, CDN), polish (mobile, accessibility, SEO), and light community features.

How will this funding be used?

Three months of me building it full time, plus what it costs to scrape and run it on real data.

• My time: $1500/month × 3 = $4500
• AI and scraping embeddings, LLM for profiles and search, scraping compute about $1,200
• Servers, database and the email service for claim-your-profile invites it's about $1,200
• Domain is about $100\

Who is on your team? What's your track record on similar projects?

Just me. Volunteers welcome if anyone is interested.

The best evidence I can build this is that I already built it. I designed and shipped the whole ATLAS demo alone: six surfaces (the map with live clustering of 1,020 people, a skill-similarity galaxy, a directory, live rooms, community rooms, conferences), a working match engine, and a full design system, deployed and running at aisa.nahdha.tech.

Background: around 4 years of applied ML and NLP engineering. MSc in Mathematical Sciences (AI for Science) at AIMS South Africa / University of Cape Town on a Google DeepMind Scholarship, thesis on neural reasoning for ARC-AGI. Research engineer on ARC-AGI at Peking University, Nov 2025 to Feb 2026. At Sultan Qaboos University I built a proposal-evaluation NLP system using multilingual embeddings scored against Oman Vision 2040 the same embedding and retrieval work the matching engine runs on. Hackathon wins: first in Qatar, first in Kigali, third at the Deep Learning Indaba. My interp work is public at github.com/AhMedDa1/mech-interp-journey.

And I'm the user. I work from Sudan, Rwanda, Oman, outside every AI safety hub, with no ready-made network. Finding mentors, collaborators and funding from here is exactly the problem this fixes. I'm building the thing I needed and couldn't find, for the people in the same spot.

What are the most likely causes and outcomes if this project fails?

Most likely ways it fails:

Nobody claims their profile. The scraper means the map is never empty, but if people don't come back to claim and edit, the data drifts. Fix: AI-drafted profiles make claiming nearly free, and it's built to be worth looking at, so people want to be on it.

Consent and privacy. Putting people on a public map is sensitive, and some people in this field keep a low profile on purpose. Opt-out and removal work from day one, profiles default to public-information-only, and anyone can take themselves off.

If it fails anyway the downside is small. The scraped public map stays up as a free reference, the code and match engine are reusable, and I'll know what makes people claim a profile and what matches they actually want, which is useful to whoever tries next.

How much money have you raised in the last 12 months, and from where?

$0 for this project. I paid for the demo and hosting myself.\