---
title: "Using LLMs to Secure Source Code"
category: "talks"
date: "2026-06-29"
time: "1:30pm-1:50pm"
track: "Security"
room: "Track 5"
speakers: ["Eugene Yan"]
sourceLabels: ["Official conference schedule", "Public YouTube metadata"]
scheduleTrack: "Security"
scheduleRoom: "Track 5"
scheduleLabels: ["Security", "Track 5", "sponsor", "confirmed"]
---
# Using LLMs to Secure Source Code

## Conference Context
- Date/time: 2026-06-29 · 1:30pm-1:50pm
- Track/room: Security · Track 5
- Speaker(s): Eugene Yan
- Session type/status: sponsor · confirmed

- Track: Security
- Room: Track 5
- Session type: sponsor
- Status: confirmed

## Session Description
Models are now finding and fixing real vulnerabilities at scale. Drawing on Anthropic's work with security teams, this talk walks a six-step workflow — threat model, sandbox, discover, verify, triage, patch — through one running example, shows where orgs actually bottleneck, and gives you a copy-paste path to your first scan.

## Media Evidence
[Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon](https://www.youtube.com/watch?v=2vlCqD6igVA) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).

- Source video: `youtube-2vlCqD6igVA`
- Slide deck: [[youtube-2vlCqD6igVA-reconstructed-slides|Reconstructed Slides: Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon]] — 2 visible slide image(s); 2 HTML recreation(s).
![[assets/reconstructed-slides/2vlCqD6igVA/slide-003.jpg]]
![[assets/reconstructed-slides/2vlCqD6igVA/slide-004.jpg]]
- Additional slide evidence: [[youtube-2vlCqD6igVA-slides|Slides: Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon]]
- Slide-derived themes for `youtube-2vlCqD6igVA`: search, query, latte, enriching, exploratory, queries, extract, catalog.

## 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-2vlCqD6igVA` — 10 slide-derived text signals
- Slide-derived themes for `youtube-2vlCqD6igVA`: search, query, latte, enriching, exploratory, queries, extract, catalog.
- Evidence links for `youtube-2vlCqD6igVA`: [[youtube-2vlCqD6igVA]], [[youtube-2vlCqD6igVA-slides]], [[youtube-2vlCqD6igVA-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
- [[eugene-yan]]

## Supporting Slides
- [[youtube-2vlCqD6igVA-slides]] — extracted from the related public AI Engineer video.

## Synthesis
### Synthesized Breakdown
# Using LLMs to Secure Source Code ## Conference Context - Date/time: 2026-06-29 · 1:30pm-1:50pm - Track/room: Security · Track 5 - Speaker(s): Eugene Yan - Session type/status: sponsor · confirmed - Track: Security - Room: Track 5 - Session type: sponsor - Status: confirmed ## Session Description Models are now finding and fixing real vulnerabilities at scale. Drawing on Anthropic's work with security teams, this talk walks a six-step workflow — threat model, sandbox, discover, verify, triage, patch — through one running example, shows where orgs actually bottleneck, and gives you a copy-paste path to your first scan. ## Media Evidence [Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon](https://www.youtube.com/watch?v=2vlCqD6igVA) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions). - Source video: `youtube-2vlCqD6igVA` - Slide deck: [[youtube-2vlCqD6igVA-reconstructed-slides|Reconstructed Slides: Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon]] — 2 visible slide image(s); 2 HTML recreation(s).

### Speaker And Company Context
- [[eugene-yan|Eugene Yan]] — Member of Technical Staff at [[anthropic|Anthropic]].

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

### Derived Links And Source Material
- [[youtube-2vlCqD6igVA]] — related YouTube source page.
- [[youtube-2vlCqD6igVA-slides]] — slide evidence.
- [[youtube-2vlCqD6igVA-reconstructed-slides]] — slide evidence.

### 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.