Filtering "Who's Hiring" with LLMs – native desktop app in Rust/egui

Name: Filtering "Who's Hiring" with LLMs – native desktop app in Rust/egui
Availability: InStock
Author: xlii

by xlii·Mar 10, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidSolve My ProblemShip It

Scores 300+ HN hiring posts against your resume, but costs ~$40 per thread.

Strengths

•Semantic matching catches 'US only' vs 'United States candidates exclusively' equivalences.
•Local Rust/egui desktop app with no backend or subscription required.
•Separate scores for evaluation, tech alignment, and compensation alignment per listing.

Weaknesses

•Gemini cache key expires after one hour requiring regeneration.
•Only Gemini Flash supported; no fallback for malformed API responses.

Post Description

Every month "Who's Hiring" drops 300+ free-form listings. Inconsistent keywords - "US only", "United States candidates exclusively", and "on-site NYC" is semantically same if you're looking for a remote job from Europe, but term-based search allows them to slip through. The thread reorders on every reload, so you lose your place.

This is a solved problem if you just throw an LLM at it. Define your requirements, attach your resume, Gemini scores and ranks everything. Built in Rust with egui.

Bring your own Gemini key — no backend, no subscription. Desktop because some listings are walls of text and a scrollable cell in a scrollable table was the only UI that made sense.