---
title: The Prime Intellect Stack
category: talks
date: '2026-06-29'
time: '4:30pm-5:30pm'
track: Workshops Day 1
room: Track 6
speakers:
  - Will Brown
sourceLabels:
  - Official conference schedule
  - Public YouTube metadata
last_auto_summarized: '2026-07-03T05:05:34.716Z'
scheduleTrack: "Workshops Day 1"
scheduleRoom: "Track 6"
scheduleLabels: ["Workshops Day 1", "Track 6", "workshop", "confirmed"]
---
# The Prime Intellect Stack

## Conference Context
- Date/time: 2026-06-29 · 4:30pm-5:30pm
- Track/room: Workshops Day 1 · Track 6
- Speaker(s): Will Brown
- Session type/status: workshop · confirmed

- Track: Workshops Day 1
- Room: Track 6
- Session type: workshop
- Status: confirmed

## Session Description
Deep dive into Prime Intellect's open-source ecosystem of post-training tools, including the verifiers and prime-rl libraries, as well as our Lab platform for self-serve training and inference.

## Media Evidence
[Training Agentic Reasoners — Will Brown, Prime Intellect](https://www.youtube.com/watch?v=PbHm2qKnu10) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).

- Source video: `youtube-PbHm2qKnu10`
- Slide deck: [[youtube-PbHm2qKnu10-dense-slides|Dense Slides: Training Agentic Reasoners — Will Brown, Prime Intellect]] — 6 visible slide image(s); 6 HTML recreation(s).
![[assets/dense-slides/PbHm2qKnu10/slide-001.jpg]]
![[assets/dense-slides/PbHm2qKnu10/slide-002.jpg]]
![[assets/dense-slides/PbHm2qKnu10/slide-003.jpg]]
- Additional slide evidence: [[youtube-PbHm2qKnu10-slides|Slides: Training Agentic Reasoners — Will Brown, Prime Intellect]], [[youtube-PbHm2qKnu10-reconstructed-slides|Reconstructed Slides: Training Agentic Reasoners — Will Brown, Prime Intellect]]
- Slide-derived themes for `youtube-PbHm2qKnu10`: kinda, works, labs, doing, four, steps, matches, performance.

- [[youtube-V-EDrhIhHzQ-transcript]] — full cached transcript markdown for the related YouTube source.

## Evidence Graph
This evidence graph is generated from currently linked source material: official schedule text, related video pages, cached transcripts, visible slide text, dense/reconstructed slide pages, and AI slide-classification audits.

### Media Signals
- `youtube-PbHm2qKnu10` — 6 slide-derived text signals
- Slide-derived themes for `youtube-PbHm2qKnu10`: kinda, works, labs, doing, four, steps, matches, performance.
- Evidence links for `youtube-PbHm2qKnu10`: [[youtube-PbHm2qKnu10]], [[youtube-PbHm2qKnu10-slides]], [[youtube-PbHm2qKnu10-dense-slides]], [[youtube-PbHm2qKnu10-reconstructed-slides]]

### Agent Reading Notes
Use these signals to refine the synopsis, topic links, people/company context, and method notes. If a source is a related external video rather than an exact official recording, keep it framed as supporting evidence.

## Summary
This workshop centers on Will Brown's Prime Intellect work on open post-training infrastructure: the `verifiers` and `prime-rl` libraries, plus the Prime Intellect Lab platform for self-serve training and inference. The connected supporting video and extracted slides, "Training Agentic Reasoners," frame the topic around building and training agentic reasoning systems rather than treating reasoning as only a prompt-engineering problem. Read the linked YouTube material as adjacent context from the same speaker and organization, not as a confirmed recording of this exact World's Fair workshop.

## Transcript Status
Related video transcript availability: English auto-captions. Treat this as supporting context, not a recording of this exact scheduled session unless later confirmed. Not fetched yet.

## People
- [[will-brown]]

## Supporting Slides
- [[youtube-PbHm2qKnu10-slides]] — extracted from the related public AI Engineer video.

## Slide Evidence
- Slide-only cropped deck: [[youtube-PbHm2qKnu10-dense-slides]] (6 viable slide images).
- Related slide/OCR pages:
- [[youtube-PbHm2qKnu10-dense-slides]]
- [[youtube-PbHm2qKnu10-reconstructed-slides]]
- [[youtube-PbHm2qKnu10-slides]]
- Slide-derived terms: `performance`, `training`, `microsoft`, `learning`, `openal`, `observed`, `same`, `compute`, `models`, `they`, `think`, `thing`, `trainer`, `graphite`, `windsurf`, `moneobb`, `mdaily`, `augment`

## Synthesis
### Synthesized Breakdown
Hey guys, how's it going? Thanks for showing up. This was a little bit of a last-minute assembly. I know a few days ago I was like talking to Swix.

### Speaker And Company Context
- No speaker profile is attached in the official schedule data.

### Topics Covered
- [[agent-security]]
- [[agentic-search]]
- [[agentic-web]]
- [[ai-sandboxes]]
- [[coding-agents]]
- [[mcp]]

### Derived Links And Source Material
- [[youtube-V-EDrhIhHzQ-transcript]] — transcript markdown; source cache `raw/sources/youtube-transcripts/V-EDrhIhHzQ.txt` (10,228 words).
- [[youtube-PbHm2qKnu10]] — related YouTube source page.
- [[youtube-PbHm2qKnu10-slides]] — slide evidence.
- [[youtube-PbHm2qKnu10-reconstructed-slides]] — slide evidence.
- [[youtube-PbHm2qKnu10-dense-slides]] — slide evidence.
- [[youtube-V-EDrhIhHzQ]] — related YouTube source page.
- [[youtube-V-EDrhIhHzQ-slides]] — slide evidence.

### Novel Concepts / Clever Methods
- [[agent-ready-accessibility|Agent-Ready Accessibility]] — Designing for agents and designing for accessibility converge around explicit structure, reachable controls, and understandable state.

### Evidence Boundary
This synthesis uses the official schedule plus cached video transcripts. Official AI Engineer World's Fair San Francisco 2026 livestreams and cut videos are primary event video sources for transcript/slide evidence; external, historical, or speaker-matched videos remain supporting context unless manually verified as exact official event recordings.

## Official YouTube Recording
- [[youtube-V-EDrhIhHzQ]] — official AI Engineer YouTube channel recording published 2026-07-13.
- Evidence status: [[youtube-V-EDrhIhHzQ-transcript]]; [[youtube-V-EDrhIhHzQ-slides]].
- Boundary: use this recording as media evidence; keep date/time/room facts tied to the official schedule.