Zara Zhang builds AI tools and writes about how people use them. Her public work includes a library of HTML presentation designs for coding agents.[1][4]
Start with Frontend Slides →
X Career
Zhang’s current About page says she is building an AI product at a stealth startup after seven years at ByteDance, where she led product marketing for enterprise software. Before that she worked as an investment analyst at GGV Capital. She studied psychology at Harvard and wrote for publications including Harvard Magazine and The Harvard Crimson. This background connects product explanation and communication with the tools she now publishes.[1][5]
Contributions
Choosing a presentation by seeing it
Frontend Slides is a skill, a set of instructions and supporting files that a coding agent can follow. It generates visual style previews, lets the user select a direction and produces a presentation as an HTML file with its CSS and JavaScript included. It can also convert an existing PowerPoint. The mechanism matters for a person who has content but cannot express a visual preference in design vocabulary: selecting a preview gives the agent a concrete target.[3]
Templates as context for an agent
Beautiful HTML Templates provides design examples for agents making slide decks. Instead of asking a model to invent every visual decision from a blank page, a template supplies a coherent structure it can adapt. The library gives an agent both the task and examples of an acceptable result, making it easier to preserve a chosen visual direction as the content changes.[4]
Tools, writing & talks
Tools
Start here
Frontend Slides
github.comA coding-agent skill that creates HTML presentations through visual previews or converts an existing PowerPoint.
Start with Frontend Slides’ beginner walkthrough and create a small deck from content you already know. No CSS expertise is required, but you need a coding agent that can read files and run the supplied workflow.
Repository owner and project author[2][3]Original source- lark-coding-agent-bridgegithub.com
Bot that links Feishu and Lark chats to a local Claude Code or Codex CLI with streaming cards.
Catalog workOriginal source - codebase-to-coursegithub.com
Claude Code skill that generates an interactive HTML course explaining how a codebase works.
Catalog workOriginal source - bilingual-subtitlesgithub.com
Claude skill that burns CJK-over-Latin bilingual subtitles into video using Pillow and ffmpeg overlay.
Catalog workOriginal source - beautiful-html-templatesgithub.com
A library of presentation designs intended for coding agents to select and adapt.
Repository owner[2][4]Original source - beautiful-feishu-whiteboardgithub.com
Gives agents 35 colour palette styles and rendering rules for editable Feishu whiteboards.
Catalog workOriginal source - call-me-skillgithub.com
Agent skill that phones you for a short interview and turns the transcript into tweet and post drafts.
Catalog workOriginal source
Apps
- youtube-to-ebookgithub.com
Python pipeline that turns YouTube channel transcripts into Claude-written EPUB ebooks, optionally emailed.
Catalog workOriginal source - youtube-digestgithub.com
Chrome side panel for YouTube with searchable transcripts, bilingual translation, AI overviews, and notes.
Catalog workOriginal source - tab-outgithub.com
Chrome extension that turns the new tab page into a dashboard of open tabs grouped by domain.
Catalog workOriginal source - reading-blockgithub.com
Chrome extension that saves articles and books a Google Calendar reading block every five saves.
Catalog workOriginal source - personalized-podcastgithub.com
Agent skill that turns content into a two-host AI podcast MP3 with editable script prompt and voices.
Catalog workOriginal source - follow-buildersgithub.com
Agent skill that sends a digest of AI builders' X posts and podcast episodes with editable summary prompts.
Catalog workOriginal source
Writing, research & talks
An article by Zara Zhang exploring how to shed school-conditioned mental blocks around learning, using language learning as a case study. It challenges the 'bad at languages' self-label and offers a reframe for adult learners returning to skills they abandoned. Written for anyone who wants to learn something new but feels held back by past academic failure.
Catalog workOriginal source