Skip to content

Rachelle Palmer

Director of Forward Deployed Engineering at MongoDB

  • 1 episode on Chain of Thought
  • 3 topics covered
  • 1 full transcript
  • Mar 2025 first on the show
Rachelle Palmer

At the time of the conversation, Rachelle Palmer was Director of Product Management at MongoDB, running a forward-deployed AI engineering team. On Chain of Thought she explained how LLMs are compressing legacy code modernization from years to months by automating documentation, test generation, and business logic conversion.

Cover art for Using AI to Modernize Your Legacy Applications | MongoDB’s Rachelle Palmer

Latest episode with Rachelle Palmer · EP 14

Using AI to Modernize Your Legacy Applications | MongoDB’s Rachelle Palmer

Rachelle Palmer, MongoDB ·

Show notes & transcript → All listening options →

Episodes

What the conversations cover

These notes summarize the episodes featuring Rachelle Palmer, including other speakers in panel conversations. Follow a source link to read the discussion in context.

EP 14 Using AI to Modernize Your Legacy Applications | MongoDB’s Rachelle Palmer

  • Rachelle runs modernization “incubators” for organizations captive to legacy code — teams that outsourced “the care and feeding” of their codebase to contractors and now find themselves “paying millions of dollars for the maintenance of their own applications” with no in-house knowledge left. Episode takeaways
  • An 80/20 split governs the work: LLMs handle roughly 80% (documentation, comments, equivalence tests) while ~20% stays “purely manual” — and the LLM share runs on a machine timescale of “minutes or hours or days” instead of the months or years a manual engineering effort would take. Episode takeaways
  • ROI is measured by an old-fashioned scoping of effort and time; sometimes a senior or staff engineer does the task manually first, then the same task is run with AI. Modernizations that would take “on the magnitude of five plus years” collapse to “months or a year,” and “the team is five people instead of 40.” Episode takeaways

In their words

Another example is your code's not commented, right? You can just have LLOs add comments to code. Another example is that you don't have test coverage of your code base. And so for us, in order to prove equivalence between a legacy application and a replacement application, you actually need to generate effectively equivalent tests between your legacy app and your new modernized app.

Rachelle Palmer Using AI to Modernize Your Legacy Applications | MongoDB’s Rachelle Palmer · 8:39

Transcript excerpts; ellipses mark omitted words. Some automatically selected quotes have filler words removed.