---
title: "I Let Agents Refactor My Codebase for 3 Weeks. Then I Read the Code."
category: "talks"
date: "2026-06-30"
time: "2:25pm-2:45pm"
track: "AI Architects: Tokenmaxxing"
room: "Leadership 2"
speakers: ["Keiji Kanazawa"]
sourceLabels: ["Official conference schedule", "Public YouTube metadata"]
scheduleTrack: "AI Architects: Tokenmaxxing"
scheduleRoom: "Leadership 2"
scheduleLabels: ["AI Architects: Tokenmaxxing", "Leadership 2", "session", "confirmed"]
---
# I Let Agents Refactor My Codebase for 3 Weeks. Then I Read the Code.

## Conference Context
- Date/time: 2026-06-30 · 2:25pm-2:45pm
- Track/room: AI Architects: Tokenmaxxing · Leadership 2
- Speaker(s): Keiji Kanazawa
- Session type/status: session · confirmed

- Track: AI Architects: Tokenmaxxing
- Room: Leadership 2
- Session type: session
- Status: confirmed

## Session Description
Lopopolo says code is a liability. Zechner got a standing ovation for "read every fucking line." I was firmly at L — letting coding agents drive a refactoring for weeks, rubber-stamping PRs, trusting the vibes. Then I actually read what they'd built and couldn't explain my own system's contracts. The interfaces weren't wrong. They were plausible. Which is worse. So I took the wheel back. But this isn't a Zechner victory lap — I'm now building better specs and evals specifically so I can move back toward L with confidence. This talk is the honest, in-progress round trip, and a framework for finding where you should sit on the spectrum today.

## Media Evidence
[AI Red Teaming Agent: Azure AI Foundry — Nagkumar Arkalgud & Keiji Kanazawa, Microsoft](https://www.youtube.com/watch?v=JhJKgRAmfIU) (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).

- Source video: `youtube-JhJKgRAmfIU`
- Slide deck: [[youtube-JhJKgRAmfIU-dense-slides|Dense Slides: AI Red Teaming Agent: Azure AI Foundry — Nagkumar Arkalgud & Keiji Kanazawa, Microsoft]] — 4 visible slide image(s); 4 HTML recreation(s).
![[assets/dense-slides/JhJKgRAmfIU/slide-001.jpg]]
![[assets/dense-slides/JhJKgRAmfIU/slide-002.jpg]]
![[assets/dense-slides/JhJKgRAmfIU/slide-003.jpg]]
- Additional slide evidence: [[youtube-JhJKgRAmfIU-slides|Slides: AI Red Teaming Agent: Azure AI Foundry — Nagkumar Arkalgud & Keiji Kanazawa, Microsoft]], [[youtube-JhJKgRAmfIU-reconstructed-slides|Reconstructed Slides: AI Red Teaming Agent: Azure AI Foundry — Nagkumar Arkalgud & Keiji Kanazawa, Microsoft]]
- Slide-derived themes for `youtube-JhJKgRAmfIU`: does, handle, loot, bank.

## 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-JhJKgRAmfIU` — 2 slide-derived text signals
- Slide-derived themes for `youtube-JhJKgRAmfIU`: does, handle, loot, bank.
- Evidence links for `youtube-JhJKgRAmfIU`: [[youtube-JhJKgRAmfIU]], [[youtube-JhJKgRAmfIU-slides]], [[youtube-JhJKgRAmfIU-dense-slides]], [[youtube-JhJKgRAmfIU-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
- [[keiji-kanazawa]]

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

## Slide Evidence
- Slide-only cropped deck: [[youtube-JhJKgRAmfIU-dense-slides]] (4 viable slide images).
- Related slide/OCR pages:
- [[youtube-JhJKgRAmfIU-dense-slides]]
- [[youtube-JhJKgRAmfIU-reconstructed-slides]]
- [[youtube-JhJKgRAmfIU-slides]]
- Slide-derived terms: `microsoft`, `azure`, `azure_ai_project_endpoint`, `foundry`, `loot`, `bank`, `initialize`, `service`, `endpoint`, `api_key`, `redteamplugin`, `problims`, `output`, `termnal`, `ports`, `node`, `smol`, `awss`

## Synthesis
### Synthesized Breakdown
# I Let Agents Refactor My Codebase for 3 Weeks. Then I Read the Code. ## Conference Context - Date/time: 2026-06-30 · 2:25pm-2:45pm - Track/room: AI Architects: Tokenmaxxing · Leadership 2 - Speaker(s): Keiji Kanazawa - Session type/status: session · confirmed - Track: AI Architects: Tokenmaxxing - Room: Leadership 2 - Session type: session - Status: confirmed ## Session Description Lopopolo says code is a liability. Zechner got a standing ovation for "read every fucking line." I was firmly at L — letting coding agents drive a refactoring for weeks, rubber-stamping PRs, trusting the vibes.

### Speaker And Company Context
- [[keiji-kanazawa|Keiji Kanazawa]] — Principal Product Manager at [[microsoft|Microsoft]].

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

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