2
3 Comments

Added AI

To solve the tedious task of drafting experience summaries, I built an AI assistant that automatically generates job, education, and project descriptions as well as highlights and keywords.

Core Features

  • 100% Free & Account-Free: Runs on a lightweight local model, eliminating subscription fees and registration steps.

  • Side-by-Side Review: Renders generated drafts in a pop-up modal so users can compare options before inserting text into the form.

System Architecture

  • Frontend: A action button sits directly beside summary fields. Clicking it bundles the record's context and triggers a preview modal.

  • Backend: A dedicated service acts as a proxy, routing requests to a local Ollama server.

  • Form Integration: The generated response feeds into the modal, allowing users to accept and auto-fill the draft with one click.

Concurrency & Scaling

Because it relies on local hardware, the server handles only one request at a time. If someone tries to generate content while it's already running, they’ll need to wait until the server becomes free. There is a "traffic signal" to indicate when the server is free. If there’s enough interest, I might offer a paid cloud version for parallel processing, but the local model handles single-user needs well.

posted toAvatar for product Resumio
Resumio
  1. 1
    The local-first approach is interesting, especially for a task like this. Are users actually hitting the single-request limit enough that parallel cloud processing would feel worth paying for?
    1. 1
      Right now - no, barely anyone using the app, but since the AI Helper is free - doesn't matter. If someone contact me and say - I need volume, probably I can implement a paid option for that. But I doubt anyone would
      1. 1
        That makes sense. The more useful signal may be whether anyone independently hits the volume constraint and asks for a way around it — that would validate the paid use case rather than assuming it.