Our Programmes
The 2026 Grassroots Internships
Twelve young people in Tanzania and Pakistan spent four weeks running community experiments with no budget, and published the findings that went against them.
Three teams, chosen by application, each took one experiment from the Grassroots Playbook and ran it in their own community between 28 July and 24 August 2026. They were paid. They were held to written integrity conditions. Three outside reviewers score the work against a guide published before any team started, and the founders take no part in that scoring.
The programme in three figures
What it cost and what it bought
Two teams in Karachi, one in Bangata Village near Arusha.
28 July to 24 August 2026, alongside school and exams.
The teams were paid a stipend; the fieldwork itself ran on no budget.
The internships are funded personally by Kubair Shirazee and his family. Peace Through Prosperity’s other programmes run on sponsorship; this one takes no grant, no institutional money and no corporate sponsor.
The teams
Three experiments, three communities
Bangata Knowledge Connectors
Five students mapped what twelve households knew and lacked, predicted in writing who should meet, made the introductions, then returned to test. They obtained the Village Executive Officer’s written authorisation unasked. Delivered with Savannah Vision Tanzania.
Scene Makers
Two students tested whether an hour a week together, phones away, changed how young people felt about focus and loneliness. Challenged on who a private racquet club excluded, they moved it to free public parks and an Edhi home.
Beyond Differences
Five students worked on inherited assumptions between students from different community backgrounds, measured comfort across those lines in two sessions, and published the check showing how much of the result rested on one participant.
The method is not ours to keep. Every experiment comes from the Grassroots Playbook, seventeen validated projects for crews of three to five, free in six languages. The Playbook is Kubair Shirazee’s, produced in partnership with Peace Through Prosperity, and sits alongside our other open material under Steal Our Stuff. Where the approach came from is set out under past programmes.
The claim
What we claim, and what we do not
Capability
Twelve young people ran a four week method and their handling of evidence measurably improved. That is fully documented for one team and partially for the other two.
Community impact
Two of the three teams took no baseline and cannot now produce one. Where a team did measure, we publish the number it got, including when the number got worse.
Capability is the aim, not a consolation prize for missing impact. A programme that leaves a result behind leaves nothing; one that leaves a method behind leaves people who can run it again without us. That is the difference between helping and enabling, and it is why fifty of the hundred marks in the guide reward how a team behaves rather than what its numbers say.
The evidence
The result got smaller every time the method got better
The first measured session. Comfort discussing ethnicity moved from 3.29 to 4.43 across seven paired records.
A tighter method and an outside facilitator, seven completers. The same measure moved from 3.71 to 4.14.
The same session with the single largest mover taken out, who alone accounts for the whole of the week four gain.
Beyond Differences measured comfort discussing ethnicity in two sessions. The figure fell each time the method tightened, and fell below zero once the team tested how much of it rested on a single participant. They keep that participant in the headline number and publish the check beside it, which is the right way round.
The week three figure carries its own range, plus 0.86 to plus 1.14, because participant codes drifted between the before and after forms and five of the eight had to be re-matched afterwards.
Governance
How the programme holds itself to account
Terms accepted before starting
Fresh work only, consent held for every named person and face, anything written after the fact marked as such, and verification rights including direct contact with participants.
A marking guide published in advance
Six criteria and one hundred marks, of which thirty go to how honestly a team handles its own evidence and twenty to whether what was published matches what was permitted. Read the scoring guide.
Outside reviewers with a licence to attack
Three reviewers from outside the programme score the work on 18 September 2026 against that guide. The founders take no part in the scoring.
The stipend is paid on verified evidence
Not on a good story. A team that measured nothing and says so scores better here than a team with a large number it cannot account for.
Why the programme is built this way is set out under our approach. The internships are the youngest of our capability programmes; Numu is the sibling that works with young community leaders over twelve weeks.
Public showcase
Friday 23 October 2026, online
12:30 London (BST) · 14:30 Arusha · 16:30 Karachi
Sixty minutes. Each team presents what it did, what changed in its community and what changed in the team itself, then takes questions. Open to guests.
To ask about a future cohort, or to run this where you are, get in touch. If you coach teams for a living and would rather give time than money, you can volunteer your Agile skills to mentor a team through its project. The rest of our work is under our programmes.