The JobFoundry Manifesto
Who this is for: anyone who needs a job and knows how to run a computer. If you can browse the web, you can use JobFoundry.
Origin: built in a real job search
JobFoundry didn't start as a startup pitch — it started because I was broke, exhausted, and struggling to find a job in a brutal market. Existing tools either locked basic features behind $30/month subscriptions, burned endless tokens reading every posting with an LLM, or produced low-quality tailoring that hallucinated fake experience.
I couldn't afford subscriptions, and I refused to apply with fabricated claims. So I started building tools to survive my own search. First came resume-ops — a CLI-first experiment in truthful, zero-token resume tailoring. But parsing needed to be much better, and the whole search cycle needed care: capturing from authenticated browser tabs, deduplicating cross-postings across 80+ boards, scoring fit locally, and tracking applications. That became JobFoundry.
Used in my own search, the pipeline was deliberately selective: let fit scoring reject most noise early, and only tailor and apply to the highest-fit roles. That discipline actually worked, and I ended up receiving multiple job offers.
I don't know whether it was the right decision or the wrong decision, but I decided to take a pause for a couple of months and put my energy into polishing this and open-sourcing it under AGPL-3.0. I wanted to see if the tools that got me through the hardest stretch of my life could help other people going through the exact same struggle.
Mass-apply automation is flooding the market: it buries strong candidates and overwhelms hiring teams. JobFoundry bets the other way — fewer, better, honest applications bring sanity back for jobseekers and employers. It is not perfect, and active development continues every day — but it is built to give jobseekers an honest edge.
1. Job hunting is hard enough. The tooling should not make it harder.
JobFoundry meets people where they already are: the browser. If you can browse job sites, you can use JobFoundry. No AI coding assistant, no terminal, no prompt skills. Install the app, add the extension, keep browsing like always — capture, dedup, scoring, tailoring, and tracking happen quietly, locally, under your control.
2. Your career data belongs to you. Entirely.
Your resume is your life on paper. JobFoundry stores everything on your machine, phones home to nobody, and works offline except for the LLM provider you choose. Local-first is not a feature here. It is the premise.
3. Reading a public job posting should never cost an LLM call.
Our zero-token philosophy: structured data comes from public ATS endpoints, feeds, and markup — free, fast, private. Models are spent only where judgment is needed: scoring fit and tailoring resumes. No AI subscription required to discover jobs.
4. Scrape from where the jobs are visible: the browser.
Server scrapers lose to Cloudflare, CAPTCHAs, and login walls daily. Your browser session walks straight through — so capture happens in your browser and the server never touches a job board. A reliability decision first, a privacy decision second.
5. A tailored resume must be true.
Tools that invent employers and skills set candidates up to fail background checks. JobFoundry re-ranks and rephrases your genuine experience under strict schema constraints, with protected fields that never change. Truth over fabrication, always.
6. The human applies. The tool prepares.
JobFoundry evaluates, drafts, organizes, and reminds. It never auto-submits, never spams employers. Fewer, better, honest applications beat spray-and-pray.
7. Free software, forever.
AGPL-3.0. Inspect every line, self-host, fork, and leave with your data anytime. Tools for vulnerable moments in people's lives must not become traps.
The canonical version lives inMANIFESTO.md.