Home / Articles / Ten days at Web Summit: what six people and around 250 agents built between 3 and 13 November, in order, and the series of articles that will tell each part
Ten days at Web Summit: what six people and around 250 agents built between 3 and 13 November, in order, and the series of articles that will tell each part
By Dinis Cruz · 2026-11-14 · article v1.3.0, 4 versions · newsroom v0.14.82 · web-summitagentspromptugalstorm-charlesteamseriescollagearticle
Abstract: Between 3 and 13 November 2026, six of us, with our agents, built more than I can fit in one article. A team operating system, an independent newsroom for Web Summit Lisbon, a site about Storm Charles that grew to more than 2,600 pages in three days, a guide to the summit's fringe that went live two and a half hours after its brief, a draft standard that rated twenty companies, four iPhone and iPad apps, an Android tablet app, and the tools to run all of it. About 20,000 commits across fifteen repositories, 224 Claude Code sessions, around 250 named agents, and more than 60 sessions working at the same time at the peak. This article is the map: the team, the sequence of events, the numbers, a collage of what we made, and the list of articles that will tell each part. I will update it as each one goes live.
In short
- Six people and around 250 agents, in ten days around Web Summit Lisbon.
- What we built: a team operating system, a newsroom, a storm site, a fringe guide, a rating standard, native apps on two platforms, and the tools to run and stop it all.
- The scale: about 20,000 commits, 224 Claude Code sessions, more than 60 working at once at the peak.
- This is the map. Each part gets its own article; the list below links them as they go live.
The team
There were six of us. Each person had their own agents, their own conductor and their own lane:
- Dinis Cruz, head conductor.
- Angelo Leggio, the graphs and live views.
- Vicky Wang, the editorial check, and the reviewer who kept the newsroom honest.
- Silvio Switha, on the ground, video, and competitive research.
- José Bota, the channels.
- Indi Gupte, communities and organisers.
The agents carried our names in theirs (role@project.person), so every commit says whose agent made it. On the Web Summit team's site alone, Angelo's agents made 895 commits, mine 853, Vicky's 767, Silvio's 320, José's 277 and Indi's 277, beside about 2,100 automatic release commits.
What happened, in order
| When | What happened |
|---|---|
| 3 Nov | MeetScribe, a Google Meet transcriber, built in a day; the book of my talk pushed to its second draft |
| 4 Nov | PrompTugal.news starts; every person gets a key for the team's site; agents get inboxes only they can write to |
| 5 Nov | The book printed as two review copies |
| 6 Nov | PrompTugal.news's busiest day: 1,249 commits |
| 7 Nov | The iPhone app grows from a 1,164-line start; the team's site has its busiest day, 699 commits |
| 8 Nov | Three iOS apps; the first App Store submission at 23:20 UTC; the Android app's first commit at 20:33 |
| 9 Nov | A signed Android build on a tablet at 00:50, four hours after the first commit; at 23:30, the Big Stop: every agent paused |
| 10 Nov | Restart at 02:40; GAMMA-Ray starts; Storm Charles reaches Lisbon and Web Summit calls off its second day at 19:09; the storm site's first commit at 21:14 |
| 11 Nov | The storm swarm: 134 sessions created in one day; Lisbon Tech Fringe live at 12:57; the Fringe iPhone app on TestFlight at 21:11; flights, transport, cloud and model-router desks; Swarm |
| 12 Nov | Shutdown at about 00:50; GAMMA-Ray stops; the hand brake at 15:15; the final hold at 18:12: "the show is over" |
| 13 Nov | The VoiceDebrief iPhone app reaches TestFlight an hour after its first commit |
The numbers
| What | How many | Where it is counted |
|---|---|---|
| Commits, main projects | about 20,000 | git history of each repository |
| Claude Code sessions created, 3 to 13 Nov | 224, 220 of them started by other sessions | the session list |
| Sessions created on 11 November alone | 134 | the session list |
| Sessions working at once, at the peak | more than 60 (64 to 67, on 11 November) | session start and last-activity times |
| Named agents | around 250 | distinct agent names in the commits |
| Storm Charles site | 2,618 pages, 1,434 versions, 3.84 billion tokens | its own accounts page |
| Native apps | 4 iOS apps (about 48,000 lines of Swift), 1 Android tablet app (about 33,000 lines of Kotlin) | the app repositories |
| Companies mapped and rated | 110 on the team's map; 20 GAMMA-Ray reports | the team's site; gamma-ray.sgit.ai |
A word on money. The dollar figures in these articles are API-equivalent: what the tokens would have cost at standard prices. We ran on flat Claude subscriptions, so they are a measure of work, not money spent.
The series
I will publish these one at a time, and link each one here as it goes live.
- Ten days at Web Summit (this article): the map.
- The team's operating system: web-summit.sgit.ai, with its board, briefs, keys, inbox lanes, vaults, company map and X-ray.
- PrompTugal.news, the newsroom we ran at Web Summit: the desks, the reviewers, the outreach drafts and the costs.
- Native apps by agents: four iOS apps from one workspace, an Android build on a tablet in four hours, the App Store and TestFlight.
- The Mac bottleneck: agents waiting three times longer than they ran.
- Stopping a hundred agents: the Big Stop, the hand brake, the final hold, and what kept running anyway.
- Agents rating agents: GAMMA-Ray, a draft standard and twenty company reports in thirty hours.
- After the storm: Storm Charles, part two: flights, transport, the cloud and the model router.
- A guide born in a storm: Lisbon Tech Fringe, live two and a half hours after its brief.
- Pennies against dollars: open-weight swarms against frontier models, and the cost of memory.
- Agents that retire themselves: 147 sessions and 106 hand-overs.
- The agents that didn't get the memo: a newsroom that ran nine days on "nothing new".
- One hour to TestFlight: VoiceDebrief, when there is something to install.
Already published, from the same ten days: Storm Charles, a newsroom built in a night, who else controls AI agents at Web Summit, an inbox only your agents can write to, a key for every person, the Web Summit team, part five and MeetScribe.
What this does not claim
- Counts depend on how you count. "Around 250 agents" is the number of distinct agent names in the commits; it includes some renamed agents. Sessions are not agents: many roles handed over to a fresh session several times.
- "More than 60 at once" is an upper bound from each session's start and last activity; a session can sit idle between turns.
- The commit total is approximate, summed across the main repositories.
- Dollar figures are API-equivalent, not money spent.
Where this comes from
The repositories behind each site, read on 13 November 2026: the Web Summit team's site, PrompTugal.news, Storm Charles, Lisbon Tech Fringe, Swarm, GAMMA-Ray, the iOS and Android apps, VoiceDebrief and MeetScribe; and the list of Claude Code sessions for the account. The sites: web-summit.sgit.ai, promptugal.news, storm-charles.promptugal.news, lisbontechfringe.com, swarm.promptugal.news, gamma-ray.sgit.ai, ios.app.promptugal.news, ios.app.voicedebrief.ai.
Peer review Not reviewed yet. Review it
Threads
Builds on
- The team's operating system: web-summit.sgit.ai, the one site where six people and their agents ran Web Summit, with a brief and a command centre for each person, a key and an inbox for each agent, the board as files, and the outside world on a map web-summit.sgit.ai: one site where six people and their agents ran Web Summit, in four layers built on git.
- PrompTugal.news, the newsroom we ran at Web Summit: desks of agents that captured, mapped, wrote and reviewed 73 signed articles in ten days, an accountant that counted every run, and fifty outreach drafts that no agent was allowed to send PrompTugal.news: 73 signed articles in ten days from desks of agents, a reviewer for every draft, and fifty outreach drafts no agent could send.
- Storm Charles, a newsroom built in a night: one person, about thirty agents and a weather face over a data room, where every figure carries its source, the time it was read, and whether it is a fact, evidence or our hypothesis Storm Charles: about thirty agents built a weather face over a data room in a night, every figure sourced, timed and labelled.
- Who else controls AI agents at Web Summit? An outreach agent mapped eleven companies in an afternoon, in their own words, and found that most of them are partners for RiskMandate as much as competitors An outreach agent mapped eleven companies that control AI agents at Web Summit; most are partners for RiskMandate as much as competitors.
- An inbox only your agents can write to: a conductor publishes one append lane per agent, each lane's token sealed to that agent's own key, so everything needed to talk privately is public and only the right agent can use it A conductor publishes one append lane per agent, each token sealed to that agent's key: a private channel built from public material.
- A key for every person: how each person's conductor issues them a key for the team's site, seals what is for them, and how a signed message turns the public web into a channel the agents can trust How each person's conductor issues them a key for the team's site: sealed to you, signed by you, and a lane the agents can trust.
- The Web Summit team, part five: the last day of setup, 180 releases, every person getting a key, an accountant counting the team's tokens, synthetic users testing the merch, and an inbox only the agents can write to The Web Summit team's last day of setup: 180 releases, keys for every person, new agents for everyone, and an inbox per agent.
- MeetScribe: transcripts and a debrief of my Google Meet calls without upgrading every Workspace account, a Chrome extension built in about two hours from a brief that would have failed without an error, and a call for testers MeetScribe: Meet transcripts and a debrief on your own OpenRouter key, built in two hours from a reviewed brief, and a call for testers.
Continued by
- Don't micromanage your agents: what changed for Vicky Wang when she stopped reading everything her agents did and started managing them, how her team was backed by agents from another team, and why every company will want specialist agents that support all its teams Vicky Wang stopped reading everything her agents did and started managing them, backed by specialist agents from another team: a model for companies.
- Atomic cells, end to end: the £500 website as AgentiCCells' first offering, an order on a queue, an agentic team that builds a public site, a customer who steers it from their browser with a key, and a handover, with no daily work for the founder AgentiCCells' first offering: a £500 website as an atomic cell, steered by the customer from their browser with a key, handed over to their GitHub.
- An agentic console: a simulated control room for a team of agentic teams, with time you can rewind, accounts drawn as trust zones, stops that reach every agent, signed messages on three channels, and the console itself as an agent An agentic console, simulated: accounts as trust zones, time you can rewind, signed messages on three channels, honest stops, and the console as an agent.
- Six setups of an agentic system: from a single task to one agent, sub-agents, several sessions, a conductor and its team, and conductors of conductors, with the altitudes of a run inside each and command and control growing at every step Six setups of an agentic system, from a task to conductors of conductors, with the altitudes of a run inside each: make the dangerous thing unreachable.
- Selling mini agentic teams: a business that sets up small agentic teams for companies, sized from one agent to a network, with a control centre to see them and to start and stop them, PrompTugal as the first customer and Agent Behaviour Policies as the most demanding version A business that sets up small agentic teams for companies, with a control centre to start and stop them; PrompTugal first, ABPs on steroids.
- A brain you arrive at: a company brain for information, decisions and insights is built for one company, one division and one team, starting with one person's and growing, and the Web Summit team already ran one A company brain is made for one company, division and team, and you arrive at it: start with a person's, then a team's. We already ran one.
- PrompTugal.news, the newsroom we ran at Web Summit: desks of agents that captured, mapped, wrote and reviewed 73 signed articles in ten days, an accountant that counted every run, and fifty outreach drafts that no agent was allowed to send PrompTugal.news: 73 signed articles in ten days from desks of agents, a reviewer for every draft, and fifty outreach drafts no agent could send.
- Always-on agents at team scale: Anthropic's five primitives for an agent that sleeps and wakes, set beside the around 250 named agents six of us ran at Web Summit, and the next problems once there are many Anthropic's five primitives for always-on agents, set beside the around 250 named agents six of us ran at Web Summit, and the next problems once there are many.
- Every hour is twenty years: the human behind Know the Reach, and why making a song and its video with AI made them more Silvio Switha's, not less Know the Reach was made with AI and by a person: Silvio Switha, whose twenty years as a musician shaped every choice. AI changed the speed, not the vision.
- Know the Reach: RiskMandate has a song on Spotify, and its music video is being made with AI, shot by shot, from about 2,300 words of human direction; this is how the first part was made RiskMandate's song Know the Reach is on Spotify; the first 53 seconds of its video were made with AI from about 2,300 words of direction.
- The team's operating system: web-summit.sgit.ai, the one site where six people and their agents ran Web Summit, with a brief and a command centre for each person, a key and an inbox for each agent, the board as files, and the outside world on a map web-summit.sgit.ai: one site where six people and their agents ran Web Summit, in four layers built on git.
- One Human, Nine Agents, Nineteen Hours is on Leanpub: the book of the day nine AI agents and I built a talk, free to download, pay what you want, and updated as it goes One Human, Nine Agents, Nineteen Hours is on Leanpub: free as a minimum, pay what you want, every update included.
Article No. 395 · start a reading list with it
Reviews
No reviews yet.