223 Pull Requests in 11 Days: Julien Dubois on Agentic Development

X postJulien DuboisResource

Julien Dubois (Java Champion and Principal Manager at Microsoft/GitHub) published a presentation playbook detailing how he used AI coding agents to build and release 'BootUI', an open-source Spring Boot 4 developer console, in just 11 days.

Key points

  1. 223 pull requests and 83,000 lines of code were merged over 11 calendar days to reach version 1.0.0 of BootUI.
  2. Dubois logged roughly 80 to 110 hours of hands-on human effort focused on steering, prompt writing, and PR reviews for GitHub Copilot agents.
  3. An equivalent solo implementation without AI was estimated at 6.5 to 8.5 months (1,100 to 1,450 hours).
  4. The deliverable includes a 5-module Maven build, an embedded Vue 3 SPA with around 40 panels, and 116 test suites across Java and Playwright.
  5. The slide deck itself was created using Copilot with reveal.js, leveraging publicly verifiable Git metadata.

Why it matters

Discussions on agentic software development often lack verifiable data. Dubois grounds his productivity claims in a public open-source repository with full Git history, offering a concrete case study on human-agent workflows.

Analysis

Dubois demonstrates a shift where the developer acts as an architect and coordinator rather than writing code directly in an IDE. By managing asynchronous agent runs, reviewing PRs, and troubleshooting CI integration issues, a single engineer could rapidly ship a multi-module application with dozens of repetitive panel implementations and tests.

Original post

Julien Dubois

@juliendubois · 9 September 2026

If you're interested in agentic development, here is the full presentation on how I'm working: t.co/Sswpgjjg1j It covers best practices, tips, pitfalls, and productivity metrics. I'll present it live (with code & demo) at selected customers and conferences like Devoxx.

196 likes5 replies396 bookmarks13.4k views

View on X
More posts in the thread (2)
  1. @OnFinality It’s in the presentation: I’m using the git logs to have exact metrics on what was done, in how much time. It’s all OSS and public, so you can check.
  2. @ChukwumaOzoeme3 GitHub Copilot App built using reveal.js -> it’s the same workflow detailled in the presentation, used to build the presentation
Selected replies (5)
  • @OnFinality @juliendubois The productivity metrics part is what I'd want to see most — how do you measure agentic gains without fooling yourself?
  • @ChukwumaOzoeme3 @juliendubois Great Julien. What did you use for the presentation ( PowerPoint?)
  • @hermallorn @juliendubois Merci
  • @shubham27907 @juliendubois Dear Sir I thank you for atleast thinking about putting all "logs" in one place. I have only seen 3 slides and had to comment already as logging has become such a painful thing and sometimes more complex and expensive than the app, itself.
  • @ModelPlaneHQ @juliendubois Curious how you measure the productivity metrics without them turning into vanity numbers. In my experience the honest signal is usually rework rate and how often the agent's plan survives first contact with the codebase.

Mentioned links

Links and tools from this entry

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Link:Julien Dubois

    223 Pull Requests in 11 Days: Building BootUI with AI Agents

    Julien Dubois presents a playbook detailing how he shipped the BootUI open-source project to version 1.0.0 in just 11 calendar days. By steering a fleet of GitHub Copilot coding agents, he merged 223 pull requests instead of spending an estimated 6.5 to 8.5 months on manual development.

    AI & LLMs· Resource

  • Repository:jdubois/boot-ui

    BootUI: Embedded Developer UI for Spring Boot and Quarkus with MCP Support

    BootUI is an embedded, local-only developer console for Spring Boot 4 and Quarkus. Alongside a Vue-based web interface, the open-source project features a CLI and a local Model Context Protocol (MCP) server that lets AI coding agents inspect runtime diagnostics during development.

    297starsJava

    AI & LLMs· Tool

  • X post:lauren

    Shipping 2,500 PRs a Month: Fully Automated CI/CD Without Human Code Reviews

    Lauren (@poteto) released a presentation originally prepared for the Cursor Compile event in London, explaining how she shipped 2,500 PRs to production in a single month. The workflow relies on AI agents that merge their own pull requests without human reviews.

    9.3kbookmarks483.7kviews

    AI & LLMs· Talk

  • Repository:github/spec-kit

    Spec Kit: Open-Source Toolkit for Spec-Driven Development with AI Agents

    GitHub's Spec Kit is an open-source framework designed for Spec-Driven Development (SDD). It helps software teams define requirements clearly before implementation, guiding AI coding agents through structured workflows.

    133.6kstarsPython

    AI & LLMs· Tool

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.