jobwatch

jobwatch

Your job hunt shouldn't be a full-time job.

For anyone applying to lots of companies: jobwatch checks the companies' own job pages for you every day, ranks the new openings against your resume, and keeps track of every application. It helps fill in forms, and you always press Submit. Free, and it runs on your own computer.

Mac / Linux
Install uvcurl -LsSf https://astral.sh/uv/install.sh | sh
orbrew install uv
Thenuv tool install jobwatch
Then runjobwatch
Windows
Install uvpowershell -c "irm https://astral.sh/uv/install.ps1 | iex"
orwinget install astral-sh.uv
Thenuv tool install jobwatch
Then runjobwatch

Works on Mac, Windows and Linux. Run jobwatch and your page opens in the browser.

Already have Python? pipx install jobwatch works too.

Use Claude Desktop? Install it with one click instead.

Twelve job boards, checked by hand.

Forty tabs and a spreadsheet that's always out of date.

The same form, typed again and again.

What it does

Watch

Watches company boards

Checks the job pages companies run themselves (Greenhouse, Lever, Workday and more), Hacker News' monthly hiring thread and investor job boards. Openings come straight from the source, with the posted pay range when there is one. It reads only what sites allow.

Rank

A short, ranked daily digest

Only new roles that match you, optionally fit-scored against your resume: must-haves met, gaps, a score out of 100.

Apply

Apply queue and form fill

A short queue of jobs to apply to, with a resume tailored per job. The Chrome extension fills the form from your details and earlier answers.

Track

Tracking through to an offer

Every stage and next step, status emails from companies as suggested updates you confirm with one click, follow-ups for applications that go quiet, a calendar feed, and offers side by side.

Prep

Interview prep

A prep sheet for each interview, answers drafted from your own stories, a mock interview and a practice video to rehearse with.

Agents

Works with your AI assistant

Ask Claude (or another assistant that supports MCP, the standard way to plug tools into AI chat) to check new jobs, queue them or prep you for a call.

See it work

The whole app in under two minutes.

Demo data: the companies and jobs shown are invented.

How it works

Six parts, each with a short video.

1

Set up

Install jobwatch and run it. It opens in your browser and asks four quick things: about you and your resume, the jobs you want, the companies to watch, and where the AI runs. Then you land on Today, your daily list of new matches.

2

Find

jobwatch watches the company boards you pick. Today ranks the new roles, optionally with a fit score against your resume.

3

Apply

A short queue of jobs to apply to, with a resume tailored per job. The Chrome extension fills application forms from your details and earlier answers. You press Submit. The extension works with the jobwatch app on your computer; install the app first.

Chrome extension
4

Track

Every stage through to an offer, with the next step and a follow-up day. Status emails from companies become suggested updates you confirm with one click, and applications that go quiet get flagged.

5

Prep

A prep sheet for each interview: what the posting asks next to your closest resume line, answers drafted from your own stories, a mock interview, and a practice video to rehearse with.

6

Compare and ask

Put your live options side by side (pay, fit, career value, company health) and see who you know at each company, so you know whom to ask for a referral.

Demo data: the companies and jobs shown are invented.

Private by design

Where the AI runs: Claude's API, Google's Antigravity (with your own subscription) or a model on your own computer; a backup if one runs out; and exactly what leaves your computer.

  • It runs on your computer. jobwatch sends nothing home.
  • Your data stays on your machine. The one exception is the AI model for fit scores and chat, if you pick a cloud one. Pick a local model and nothing leaves at all.
  • You press Submit yourself. jobwatch finds and ranks. It never applies for you.
Want the AI to run on your own computer? Pick a model by how much memory (RAM) you have.

Sized by how much memory your computer has. jobwatch's page suggests one for yours.

MemoryMac with Apple Silicon (MLX, runs on the GPU via Metal)Windows, Linux or Intel Mac (Ollama)Also good: Gemma 4 (MLX first, then Ollama)
8 GB Qwen3.5-4B, 4-bit
Qwen3.5-4B-MLX-4bit ↗suggested, sized by memory
qwen3.5:4b-q4_K_M ↗ gemma-4-e2b-it-4bit ↗
gemma4:e2b-it-q4_K_M ↗
16 GB
tested on a 16 GB Mac
Qwen3.5-9B, 4-bit
Qwen3.5-9B-MLX-4bit ↗
qwen3.5:9b-q4_K_M ↗same model; the test ran the Apple Silicon build gemma-4-12B-it-4bit ↗
gemma4:12b-it-q4_K_M ↗tested: most accurate (29/29 vs 28/29), about 50% slower, +1.7 GB
24-32 GB Qwen3.8-27B, 4-bit
Qwen3.8-27B-4bit ↗suggested, sized by memory
qwen3.8:27b-q4_K_M ↗ gemma-4-26b-a4b-it-4bit ↗
gemma4:26b-a4b-it-q4_K_M ↗
48 GB and up Qwen3.6-35B-A3B, 4-bit
Qwen3.6-35B-A3B-4bit ↗
8-bit with 64 GB or more:
Qwen3.6-35B-A3B-8bit ↗suggested, sized by memory
qwen3.6:35b-a3b-q4_K_M ↗
8-bit with 64 GB or more:
qwen3.6:35b-a3b-q8_0 ↗
gemma-4-31b-it-4bit ↗
gemma4:31b-it-q4_K_M ↗

On computers with 48 GB or more, 35B-A3B-style models (about 3B active parameters) are the fastest choice.

On an Apple Silicon Mac, use MLX: it's built for Apple's chips and runs faster than Ollama there. Ollama is for computers MLX doesn't run on.

We tested only the 16 GB row: it matched 28 of 29 hand-checked scores, at about 27 seconds a job. The other rows are suggested, sized by memory. The Gemma 4 column is an alternative for each row; on Apple Silicon it is served the same way as the Qwen beside it.

Install it

Mac / Linux
Install uvcurl -LsSf https://astral.sh/uv/install.sh | sh
orbrew install uv
Thenuv tool install jobwatch
Then runjobwatch
Windows
Install uvpowershell -c "irm https://astral.sh/uv/install.ps1 | iex"
orwinget install astral-sh.uv
Thenuv tool install jobwatch
Then runjobwatch

Already have Python? pipx install jobwatch works too.

Read more

Join in

Star it on GitHub

jobwatch collects no usage data, so a star is how I know it helped.

Star on GitHub

Issues and PRs welcome

Found a bug, a board that won't load, or want to add something? Open an issue or send a pull request.

Issues · The repo

Buy me a coffee

If it landed you an interview, a coffee says thanks.

Buy a coffee