---
title: "Special topics in Kernels, RL, Reward Hacking in Agents"
category: "talks"
date: "2026-06-29"
time: "2:20pm-5:30pm"
track: "Workshops Day 1"
room: "Track 3"
speakers: ["Daniel Han"]
sourceLabels: ["Official conference schedule", "Public YouTube metadata"]
scheduleTrack: "Workshops Day 1"
scheduleRoom: "Track 3"
scheduleLabels: ["Workshops Day 1", "Track 3", "session", "confirmed"]
---
# Special topics in Kernels, RL, Reward Hacking in Agents

## Conference Context
- Date/time: 2026-06-29 · 2:20pm-5:30pm
- Track/room: Workshops Day 1 · Track 3
- Speaker(s): Daniel Han
- Session type/status: session · confirmed

- Track: Workshops Day 1
- Room: Track 3
- Session type: session
- Status: confirmed

## Session Description
An advanced seminar (good prerequisites: Daniel's 2024 and 2025 hit AIE workshops, but all are welcome!) PLS WATCH: https://www.youtube.com/@aiDotEngineer/search?query=daniel%20han

## Media Evidence
[(Full Workshop) Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han](https://www.youtube.com/watch?v=OkEGJ5G3foU) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).

- Source video: `youtube-OkEGJ5G3foU`
- Slide deck: [[youtube-OkEGJ5G3foU-dense-slides|Dense Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]] — no readable content slides after AI classification.
- Additional slide evidence: [[youtube-OkEGJ5G3foU-slides|Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]], [[youtube-OkEGJ5G3foU-reconstructed-slides|Reconstructed Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]]
- Slide-derived themes for `youtube-OkEGJ5G3foU`: fixes, chat, template, multiple, llama, research, google, github.

## 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-OkEGJ5G3foU` — 3 slide-derived text signals
- Slide-derived themes for `youtube-OkEGJ5G3foU`: fixes, chat, template, multiple, llama, research, google, github.
- Evidence links for `youtube-OkEGJ5G3foU`: [[youtube-OkEGJ5G3foU]], [[youtube-OkEGJ5G3foU-slides]], [[youtube-OkEGJ5G3foU-dense-slides]], [[youtube-OkEGJ5G3foU-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.

## 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
- [[daniel-han]]

## Supporting Slides
- [[youtube-OkEGJ5G3foU-slides]] — extracted from the related public AI Engineer video.

## Slide Evidence
- Slide-only cropped deck: [[youtube-OkEGJ5G3foU-dense-slides]] (1 viable slide images).
- Related slide/OCR pages:
- [[youtube-OkEGJ5G3foU-dense-slides]]
- [[youtube-OkEGJ5G3foU-reconstructed-slides]]
- [[youtube-OkEGJ5G3foU-slides]]
- Slide-derived terms: `microsoft`, `contributions`, `fixes`, `awss`, `graphite`, `windsurf`, `mongobr`, `mdaily`, `augment`, `code`, `workos`, `unsloth`, `daniel`, `deep-dive`, `kernels`, `quantization`, `aaeoat`, `naan`

## Synthesis
### Synthesized Breakdown
# Special topics in Kernels, RL, Reward Hacking in Agents ## Conference Context - Date/time: 2026-06-29 · 2:20pm-5:30pm - Track/room: Workshops Day 1 · Track 3 - Speaker(s): Daniel Han - Session type/status: session · confirmed - Track: Workshops Day 1 - Room: Track 3 - Session type: session - Status: confirmed ## Session Description An advanced seminar (good prerequisites: Daniel's 2024 and 2025 hit AIE workshops, but all are welcome!) PLS WATCH: https://www.youtube.com/@aiDotEngineer/search?query=daniel%20han ## Media Evidence [(Full Workshop) Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han](https://www.youtube.com/watch?v=OkEGJ5G3foU) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions). - Source video: `youtube-OkEGJ5G3foU` - Slide deck: [[youtube-OkEGJ5G3foU-dense-slides|Dense Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]] — no readable content slides after AI classification. - Additional slide evidence: [[youtube-OkEGJ5G3foU-slides|Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]], [[youtube-OkEGJ5G3foU-reconstructed-slides|Reconstructed Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han]] - Slide-derived themes for `youtube-OkEGJ5G3foU`: fixes, chat, template, multiple, llama, research, google, github. ## 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.

### Speaker And Company Context
- [[daniel-han|Daniel Han]] — Co-founder at [[unsloth|Unsloth]].

### Topics Covered
- [[agentic-search]]
- [[coding-agents]]

### Derived Links And Source Material
- [[youtube-OkEGJ5G3foU]] — related YouTube source page.
- [[youtube-OkEGJ5G3foU-slides]] — slide evidence.
- [[youtube-OkEGJ5G3foU-reconstructed-slides]] — slide evidence.
- [[youtube-OkEGJ5G3foU-dense-slides]] — slide evidence.
- [[youtube-uIiA6DquRiE]] — related YouTube source page.

### Novel Concepts / Clever Methods
- No highlighted novel concept has been detected yet.

### Evidence Boundary
This synthesis is based on the official schedule and linked source pages. It should be revisited when exact session recordings or transcript-backed secondary sources are available.

## Official YouTube Recording
- [[youtube-uIiA6DquRiE]] — scheduled official AI Engineer YouTube premiere for 2026-07-17.
- Evidence status: transcript/slide enrichment pending.
- Boundary: use this recording as media evidence; keep date/time/room facts tied to the official schedule.