Global Health Project Sprint — Participant Info Pack

[Apply Now] | Application deadline: Friday 18 September | Runs 5 Oct – 4 Dec 2026

A 6-week project sprint for people building high-impact careers in global health

Run jointly by Effective Altruism South Africa, Effective Altruism Nigeria, and Health Progress Hub.

Finish something. Six weeks, one real problem, a mentor who knows the field, and a published piece of work with your name on it that anyone can read.

🎯 At a Glance

💭 What is this?

If you've done a fellowship, a course, or your own reading, you probably hold a set of frameworks: burden metrics, cost-effectiveness reasoning, governance mapping, implementation drivers, advocacy strategy, without having had the chance to use any of them on a problem where the answer wasn't already known.

A few weeks of reading is a beginning, not a finish. It can introduce a field; it can't take you from understanding global health and development to being trusted with a piece of work in it. This sprint is the next step: you pick a problem, work it for seven weeks with a mentor, and produce something finished that gets published under your name.

Two things happen when you do that. You learn the frameworks properly, because applying one to real, incomplete, awkward data teaches more than any session about it can, and it shows you where your knowledge runs out, which is what makes further learning targeted instead of vague. And you end up with something a hiring manager, funder, or programme lead can actually read, instead of a certificate that asks them to take your word for it.

Open-access resources for the global health ecosystem rather than specific deliverables for organisations. The aim is high-quality, independent resources and tools the whole global health community can use, reference, or expand on. Each project stands on its own, it is complete without anyone adopting it. Where a finished deliverable does open a conversation with an organisation, we'll support that.

🛤️ Choose your track

Track B — the four extra requirements

If you propose your own project, we're asking you to establish what the menu options already have established: that the work is worth doing and can be done. Specifically:

1. Problem & Core Question: Define the specific issue you’re tackling and the primary question your project will answer.
2. The Business Case (ITN): Explain why this matters, why it remains unsolved, and who else works in this space. Clarify why a small, unbudgeted team can meaningfully address it in seven weeks.
3. Deliverable Specifications: Describe the final product in concrete detail—including its length, file format, and where it will be stored or published.
4. Target User: Identify who would actually use this output and what action they would take with it. Focus on a realistic target user rather than securing a formal commitment right now
5. Data & Methodology: Outline your approach and list 3–5 specific data sources, adding a quick note for each on whether the data exists and how you'll access it.
6. Team & Ownership: List team members and clear responsibilities for each person.
7. Timeline: Provide a week-by-week roadmap with clear milestones across the seven-week period.

If your Track B proposal isn't accepted, you'll be offered a place on a menu project instead — you don't drop out of the phase.

🚀 Why do this

You'll have something to show. Applications at impact-focused organisations keep asking the same question: describe your most impressive project. If you have nothing, you have nothing to write in that box — and it's the question that separates candidates with similar CVs.

Published work is checkable. A private document asks someone to trust you. A public URL with your name on it lets them look. People who hire researchers repeatedly name independent published work as one of the highest-value things an applicant can do, precisely because it lets someone assess your reasoning directly.

It's the cheapest way to find out if you want this work. 80,000 Hours describes a ladder of cheap tests — reading about a field, then talking to people in it, then doing a short project, then committing to something long and expensive like graduate study. Seven weeks is a much cheaper test than a master's degree, and it's worth doing even if the deliverable never lands you a role.

You'll actually learn the material. This is the part people underrate. A framework you've applied once, badly, with someone correcting you, is worth more than three you can describe.

Structure you can't easily build alone. A submission deadline, a mentor to review your work weekly. Written feedback against a success rubric, a team, or an accountability buddy to work with. A guaranteed reader: EASA, EAN, and HPH each read every finished deliverable and respond in writing — you will not be left wondering what happened to your work.

📋 The project menuEach project is pre-scoped, each has a written case for why the problem is important, neglected and tractable.

Full descriptions — deliverable, milestones, mentor profile — are in the project menu document.

Projects don't have to map onto a fellowship framework. What every project must do is make a real case that its problem is important, neglected, and tractable anywhere across global health and development. The two are inseparable in practice, and the menu reflects that.

📖 How the six weeks work

You'll meet your mentor weekly for 45–60 minutes. Every session has something due not a micro management tool, but built in to help you deliver your project at submission deadline.

Send your work 24 hours before each session. If you don't, the meeting becomes a 15-minute check-in rather than a review — which wastes your mentor's time and yours.

Descoping at week 3 is a legitimate, good outcome. A smaller project done properly scores higher than an ambitious one done partly. We mean that, and the rubric reflects it.

🚧 What's in scope

Projects are research, advocacy, and scoping — and that includes real contact with the world, which is often what makes a project good rather than academic. Stakeholder engagement sessions, interviews with officials and practitioners, surveys of professionals, co-design workshops, and feasibility scoping with a host organisation are all encouraged, and need no extra process.

What's out of scope is implementation with real populations: anything involving patients, minors, or people in crisis; collecting personal health data; offering health advice, screening or referral; distributing anything; or community pilots that create a service expectation. Anything needing ethics or regulatory clearance you can't obtain inside seven weeks is also out.

If your project sits somewhere in between — collecting data from members of the public, say, or a small non-clinical pilot inside an organisation's own operations — flag it in your application. It may well be fine, but it needs a named accountable person at the host organisation, mentor sign-off on the design, a consent and data-handling plan compliant with POPIA or NDPR, and a plan for what happens at week 7.

The reason for the boundary: when a research project falls apart you lose seven weeks; when a delivery project falls apart someone else loses far more.

✅ What "finished" means

A team at level 1–2 by the week-three midpoint is the signal worth a mentor conversation — not attendance or engagement scores.

Work that meets the publication bar goes live on the EASA, EAN, or HPH platforms with your name on it. Work that doesn't get the score, written feedback, and an explanation — non-publication is a legitimate outcome, and we'd rather tell you honestly than publish something under three organisations' names before it's ready.

Certificates

You get one if you submit by the deadline and reach level 2 or above. Late submissions within the grace window get full feedback and a participation certificate, and can still be published later if the work clears the bar — but not a showcase slot.

🚫 What this is not

We'd rather be clear now than have you disappointed in December.

• Not a job. No role, internship, or introduction is guaranteed.
• Not funded for participants. No microgrants or stipends this round. (Mentors receive a stipend.)
• Not implementation experience. See the scope section.
• Not a qualification. It produces evidence, not credentials.
• Not a promise anyone adopts your work. What we do promise: all three organisations read every finished deliverable and respond in writing, and work meeting the bar gets published with your name on it.

📅 Key dates

❓ FAQ

Do I need to have done a previous fellowship? No. Alumni of the Global Health Introductory Fellowship and the Introduction to Effective Altruism Program are especially well placed, but the sprint is open to anyone interested in high-impact global health work who has some grounding in ideas like importance, tractability, neglectedness, and counterfactual impact.

I don't have a team. Can I still apply? Yes. We'll help you find one, or apply solo — solo projects are a real option, not a fallback. You'll be paired with another solo participant for peer accountability.

Is solo actually harder? Different. You lose the redundancy of a team, so a bad fortnight hurts more. But you gain about an hour a week back — no team meeting — and you never wait on anyone. Solo proposals are held to a tighter scope, which is a help rather than a penalty.

Can I propose my own project on my own? Yes. Track B is open to solo applicants; the four evidence requirements do the filtering, not a rule about team size.

How much time will this really take? 5–10 hours a week is honest for a menu project at the stated scope. It goes over when scope creeps, which is exactly what the week-1 and week-4 gates exist to prevent. Tell your mentor early if it's running hot.

What if I fall behind? Say so at the week-3 gate and descope. A smaller project finished well scores better than an ambitious one abandoned. This is the single most common thing we'd like you to do and the one people avoid.

What if my data doesn't exist? That's what week 2 is for. You'll descope or redirect, and that's a normal outcome rather than a failure — you'll have learned something most people learn much later and more expensively.

Do I need a partner organisation? No, and you're not expected to find one. Your project is complete as published work.

Will it definitely be published? If it meets the bar, yes. If it doesn't, you get the score, the written feedback, and a certificate. We'd rather tell you honestly than publish work under three organisations' names that isn't ready.

How competitive is it? We expect more applications than places. The strongest applications show a realistic scope, a clear case that the problem matters and is addressable, and data you've actually checked exists.

Does my project have to be about a fellowship topic? No. It has to be about a problem you can show is neglected and tractable, anywhere across global health and development.

I've never done research before. Neither have most applicants. Projects 6 and 4 are the most accessible starting points, and your mentor is there for exactly this.

☀️ About the three organisations

Effective Altruism South Africa

Supports people to maximise the impact of their resources and careers, through capacity-building programmes, career advising, and community. Since 2024: 160+ programme completions, ~24 high-impact roles accepted, ~5 high-impact initiatives launched.

Website Facebook LinkedIn Instagram Youtube Mailing list

Email: hello@effectivealtruism.co.za

Effective Altruism Nigeria

Builds and supports the community of people in Nigeria working on the world's most pressing problems.

Website Facebook LinkedInEmail: info@eanigeria.org

Health Progress Hub

Works on research and advocacy for global health progress, with a focus on neglected problems and African contexts.

Website LinkedIn SubstackEmail: info@healthprogresshub.org

✍️ How to apply

[Apply Now] | Deadline: Friday 18 September 2026

One form for both tracks. Following the proposal guideline, it asks for: the problem and your core question; the case for why it matters, why it's unsolved, and why a small unbudgeted team can move it in seven weeks; the deliverable in concrete detail (length, format, where it lives); your target user and what they'd do with it; your method plus 3–5 named data sources with a note on access; team members and who owns what; and a week-by-week timeline. Around 45 minutes for Track A, longer for Track B.

Not sure whether to apply? Apply. Come to an office hour first if it helps. We'd rather help you work out whether it's a fit than have you rule yourself out.

Questions: hello@effectivealtruism.co.za