The Missing Layer: Design Taste in AI Agents // Stop Letting Your Agents Ship Ugly UIs
Conference Context
- Date/time: 2026-06-30 · 2:50pm-3:10pm
- Track/room: Design Engineering · Track 6
- Speaker(s): Hassan El Mghari
- Session type/status: session · confirmed
- Track: Design Engineering
- Room: Track 6
- Session type: session
- Status: confirmed
Session Description
Alt titles: "UI Looksmaxxing for Agents", "Teaching agents design taste", or "How to give your agents great design taste". I've been exploring how to give coding agents good design taste for the last few months. In this talk, I'm going to go over how to help your agents give you UIs that don't suck and that look quite good out of the box. The key is giving them enough context in what you're building + real inspiration in the form of screenshots. I'll also go over an upcoming design taste OSS project I'm working on (harness-agnostic + will ship with a prompt builder, MCP server w/ inspo, and a design eng skill) & talk about how to I use it to build my apps.
Media Evidence
Using OSS models to build AI apps with millions of users — Hassan El Mghari (speaker-match related prior/adjacent AI Engineer video; captions: English auto-captions).
- Source video:
youtube-gcseUQJ6Gbg - Slide deck: Slides: Using OSS models to build AI apps with millions of users — Hassan El Mghari — 1 visible slide image(s); 1 HTML recreation(s).
- Slide-derived themes for
youtube-gcseUQJ6Gbg: section, agenda, introduction, together, demos, architecture, typical, tech.
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-gcseUQJ6Gbg— 7 slide-derived text signals- Slide-derived themes for
youtube-gcseUQJ6Gbg: section, agenda, introduction, together, demos, architecture, typical, tech. - Evidence links for
youtube-gcseUQJ6Gbg: youtube gcseUQJ6Gbg, youtube gcseUQJ6Gbg 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
Supporting Slides
- youtube gcseUQJ6Gbg slides — extracted from the related public AI Engineer video.
Attendance Visibility
No high-confidence attendance icon signal is shown for this talk. The sampled video evidence was either low confidence, source-proxy-only, or did not expose a clear audience view.
Synthesis
Synthesized Breakdown
The Missing Layer: Design Taste in AI Agents // Stop Letting Your Agents Ship Ugly UIs ## Conference Context - Date/time: 2026-06-30 · 2:50pm-3:10pm - Track/room: Design Engineering · Track 6 - Speaker(s): Hassan El Mghari - Session type/status: session · confirmed - Track: Design Engineering - Room: Track 6 - Session type: session - Status: confirmed ## Session Description Alt titles: "UI Looksmaxxing for Agents", "Teaching agents design taste", or "How to give your agents great design taste". I've been exploring how to give coding agents good design taste for the last few months. In this talk, I'm going to go over how to help your agents give you UIs that don't suck and that look quite good out of the box. The key is giving them enough context in what you're building + real inspiration in the form of screenshots.
Speaker And Company Context
- Hassan El Mghari — Director of Developer Experience at Together AI.
Topics Covered
Derived Links And Source Material
- youtube gcseUQJ6Gbg — related YouTube source page.
- youtube gcseUQJ6Gbg 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.