Running - sameernagar-hub/regen-engine GitHub Wiki
python -m engine scan 1 # every registered board, postings from the last 24 h
python -m engine newgrad 1 # new-grad leads resolved to employer ATS boards
python -m engine batch b1 <id,id,...> # fit gate + one tailored resume per job
REGEN_TABS=3 python -m engine apply batches/b1.json
python -m engine report # verified submissions only (proof on disk)Settings: REGEN_TABS (tabs in the round-robin, default 3), REGEN_PW_PROFILE (browser profile folder, so a second
applier can run next to the first), REGEN_MAX_PER_DAY (daily cap, default 25), REGEN_DRAFT=0 (send open-ended
questions to you instead of drafting).
- A job reaching the code step writes
workspace/codes/<board>_<id>.wait(company and form URL) and keeps waiting while the other tabs work. - In a signed-in Gmail tab, run
engine/discovery/gmail_codes.js; it returns[{company, code}]from the search rows without opening any message. -
python -m engine codes '<that json>'writes each code only to the job whose company the email names.
Save recruiting emails as [{id, date, from, subject, snippet}] and run python -m engine inbox <file>, or use
python -m engine inbox --imap 3 with an app password in .env. Then python -m engine learn.
-
workspace/human_queue.mdand the live view's "waiting on you" list each job's unanswered questions. - With
REGEN_API_WRITE=1on the local API, answer them in the page. Answers go toprofile/answers.jsonwith their source and are reused; the job lands inbatches/requeue.json.
Create workspace/STOP. The applier stops before the next job; delete the file to resume.
python -m pytest -q --cov=engine --cov=apps/api # unit, API and headless form tests
python scripts/bench_scheduler.py 20 3 # scheduler speed-up (deterministic)
python scripts/bench_tailor.py <old tailor.py> # identical output + speed vs an older versionRelated: Home · Architecture · Discovery · Fit gate · Tailoring · Apply engine · Email codes · Outcomes · Platform · Metrics