---
title: "Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han"
category: "slides"
video_id: "OkEGJ5G3foU"
sourceLabels: ["Public YouTube video frames", "Public YouTube metadata"]
---

# Slides: [Full Workshop] Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han

## Source Video
[(Full Workshop) Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han](https://www.youtube.com/watch?v=OkEGJ5G3foU)

## Relationship To World's Fair 2026
These slides are extracted from a public AI Engineer YouTube video connected to World's Fair 2026. Speaker-matched clips are supporting context unless later confirmed as exact session recordings; official livestream recordings are day-level/event-level source material.

## Related Scheduled Sessions
- No individual scheduled session mapping has been assigned yet; treat this as an event livestream deck.

## Extracted Slides
![[assets/slides/OkEGJ5G3foU/slide-001.jpg]]

OCR text:

> INNOVATIONPARTNER
> aws
> PLATINUMSPONSORS
> Graphite
> WWindsurf
> MongoDB
> daily
> augment code
> Workos

![[assets/slides/OkEGJ5G3foU/slide-002.jpg]]

OCR text:

> © unsloth Daniel Han
> Deep-Dive: RL,
> Kernels, Agents & ; |
> Quantization i oa aaeoat
> Naan eect
> Ga.@)
> sot,
> , aws
> | wnat, | ee

![[assets/slides/OkEGJ5G3foU/slide-003.jpg]]

OCR text:

> I OPEN source
> 
> 5 ing van ited dn Bucecllnasd ng solng nike contributions
> 
> [se ong] Sufhees during my experiments wth Pre-4 the ofner day it tums out
> 
> the Pri-4 Sokenizer Getinmon as released Dy Mcrosott Nad a bu i it, and there
> 
> wat a imal bug in the chat template at vrei Phi-3 + 4 Bug fixes
> 
> Oaniel s cow t UF Mes i
> 
> oe chaise sien fan PG a 9 Ma Ne Banana SF Merged by Microsoft
> 
> Posted 11°" January 2079 at 1 20 am
> Multiple Llama 4 fixes
> and contributions to
> llama.cpp
> Recently worked with
> Quwen team on Qwen3
> and Mistral on Devstral
> 
> ©) unsioth
> \ >,

![[assets/slides/OkEGJ5G3foU/slide-004.jpg]]

OCR text:

> env:
> READY!

![[assets/slides/OkEGJ5G3foU/slide-005.jpg]]

OCR text:

> AIEUnsloth-GoogleSlides
> GoogleColab
> https://colab.research.google.com/github/
> AIE
> Loading...
> GoogleColab-Mc
> 10:56AM
> Microsoft
> smol?
> Word'sFair

![[assets/slides/OkEGJ5G3foU/slide-006.jpg]]

OCR text:

> AIE TYPE_CHECKING':False, 'bundle_triton_into_fx_graph_cache':True, cnncache_precompile_headers':True, inplace_padding':True, can_inplace_pad_graph_input':False, disable_progress':True, verbose_progress':False, fx_graph_remote_cache':None, bundled_autotune_remote_cache':None, sleep_Sec_TESTING_ONLY':None, enable_auto_functionalized_v2':True, fx_graph_cache':True, autotune_local_cache':True, force_disable_caches':False, custom_op_default_layout_constraint':'needs_fixed_stride_order', triton_kernel_default_layout_constraint':'needs_fixed_stride_order' debug':False, autotune_remote_cache':None, cpp_wrapper':False, softmax':True,
> Microsoft smol?
> Worid'sFair

![[assets/slides/OkEGJ5G3foU/slide-007.jpg]]

OCR text:

> AI Engineer
> World's Fair

## Slide-Derived Subjects To Review
Subject extraction uses video title, related session titles/descriptions, transcript context, and OCR text when available. OCR is best-effort and should be reviewed against the embedded slide images.