Skip to content
agentusehub
Sign in+ Submit
Software & DevOpsAny

Reverse engineer legacy systems without source code

Thoughtworks consultants used Gemini 2.5 Pro in a two-week pilot to produce validated blueprints of legacy systems' functionality.

Read the experience
Time saved
Not reported
Setup time
Not reported
Running cost
Not reported
Cost range
Not reported

Use case overview

Thoughtworks consultants used Gemini 2.5 Pro in a two-week pilot to produce validated blueprints of legacy systems' functionality.

Read the original report ↗

Reported experience

  1. What they did

    Thoughtworks consultants used Gemini 2.5 Pro to reverse engineer legacy systems that had no source code available. The AI helped extract what the systems do, and the output was validated "blueprints" of their functionality. This was a two-week pilot. The listing I read is only a short blurb, so it gives no detail on inputs or on the steps people took.

  2. How it went

    The pilot produced validated functionality blueprints in two weeks. The blurb says it showed AI could cut the time and risk of modernizing opaque systems, with speed having to be balanced against validation.

  3. Worth knowing

    Validation is the catch. Speed has to be balanced against checking the AI's output. The full InfoQ article, not this category listing, has the details.

Example prompts

Try with Gemini
I have a legacy system in [folder path] and no documentation. Read through the code, configs, and database schemas and write a blueprint in a markdown file covering what the system does, its main workflows, data flows, and business rules. Mark anything you are unsure about as unverified. Do not modify any files other than the blueprint. Stop when every major module is described.

Related use cases

Software & DevOps

Port a Chrome extension to Safari

The author had ChatGPT take the Session Buddy extension source from Chrome, port it to Safari and install it, taking about 20 minutes.

ChatGPT