apply-agent 👷💭 ^noname1+_wip

Self-hosted job scraper runner, with self-hosted LLM-powered CV matching.

Caution

🍪 It’s possible to filter out legitimate jobs, so use it with caution.

What’s automated

There are three main automated processes. They can run in parallel.

1. Get Jobs	2. Filter Out the Noise	3. Evaluate
☑️ Visit job site(s)	☑️ Normalise scraped jobs	☑️ Challenge shortlisted jobs
☑️ Search by pre-defined criteria	☑️ Run batch scoring and find signals	☑️ Put them into the state machine
☑️ Download results	☑️ Screen out irrelevant jobs	🔘 Enjoy approved jobs

If something is ambiguous during the job categorisation process, the user (HITL) will need to answer a few questions to clarify it.

How’s it going?

Batch reject:
“Not worth thinking about.”

Agent reject:
“Thought about it carefully and decided no.”

How to run?

The simplest way to run the project is with Docker Compose. It automatically sets up local LLM models (CPU-only for now) and starts searching for jobs right away. You’ll need about 3 GB of disk space with the default settings. Clone the repository and run with the default settings:

1. Clone the repository

2. Run Docker Compose

   docker compose up

Documentation

Go to the index page (already legacy ).

Footnotes

1. Apply in the repo name is confusing — it doesn’t actually do anything. ↩