G / 01GIXTS OBSERVATORY
LOCAL TIMELOCAL SESSION

The toolkit / Résumé

AI résumé enhancer and builder, with your own model

Rewrite bullets, tailor to a posting, or generate a whole résumé, using your own AI key.

Available in VultureFree, local profileNo watermark

The overview

A little less work. A better starting point.

Point the app at OpenAI, Anthropic, Gemini, a local Ollama model or the built-in WebGPU model, and it will sharpen a bullet, tailor a whole résumé to a posting or write one from scratch. Your key, your model, your bill: there is no per-token markup and no middleman.

01 / Capabilities

What it does

  • Per-bullet enhancement

    Rewrite one line at a time with the surrounding role as context, so the result stays true to what you actually did rather than inventing a career.

  • Tailored to one posting

    Job Helper reads a job description and returns a tailored résumé, a cover letter, a company brief and a preparation to-do list in a single pass.

  • Grounded in the analysis

    The coach is given the real score breakdown and keyword gaps, not just the text, so its advice matches the report rather than contradicting it.

  • Any model, including local

    OpenAI, Anthropic, Gemini, OpenRouter, a local Ollama server or a WebGPU model running in the browser. Nothing is hard-wired to one vendor.

  • Humanise afterwards

    Anything a model writes can be run straight through the AI check and rewritten in plainer language until it reads like you again.

02 / The workflow

How to use it

  1. Add your own API key in Settings, or start an Ollama server, or pick the built-in browser model.
  2. Open a résumé and use Enhance on a single bullet to see the tone before committing to more.
  3. Paste a posting into Job Helper for a tailored résumé, cover letter and prep list in one run.
  4. Run the AI check over the result and swap out any phrasing that reads as generated.
  5. Score against the posting, fix the remaining gaps and export.

03 / Good to know

Questions, answered.

Do I have to pay for AI?

Only your model provider, and only if you choose a paid one. A local Ollama model or the built-in WebGPU model costs nothing. Vulture Pro is an optional way to use our keys instead of your own.

Does my résumé go to the model provider?

Only the text you send, and only to the provider you configured. Vulture adds no server in between when you use your own key.

Will it invent experience?

The prompts are written to rewrite rather than fabricate, and everything is editable before it is saved. Read what it produces: it is a draft, not a submission.

Which model should I use?

For rewriting, almost any current model is enough. For tailoring a whole résumé to a posting, a larger model is noticeably better.

Better together

More free tools

The rest of the set