[Feature]: Profile wizard: parse work-experience blocks from resumes #8

What problem does this solve?

The profile wizard at agent/profile_wizard.py already extracts name, email, phone, school, degree, and skills from a resume PDF. But the resulting data/master-resume.json ships with experience: [] — the user's actual work history isn't parsed.

This blocks resume tailoring (the loop's tailor.py re-ranks bullets per JD, but only if there are bullets to re-rank).

What to do

1. Open agent/resume_parser.py. The text extraction (extract_resume_text) already returns plain text from PDF/DOCX.
2. Add an _extract_experience() function that finds work-history blocks. Heuristics that work:
   • Lines matching ^(.+?)\s+(\d{4}.*?(present|\d{4})) are likely role/dates
   • Bullets starting with •, -, or * under a role are bullets for that role
   • Section headers like "EXPERIENCE", "WORK EXPERIENCE", "PROFESSIONAL EXPERIENCE" mark the boundary
3. Return a list of {"company": str, "title": str, "start": str, "end": str, "bullets": [str]} dicts.
4. Wire it into profile_wizard.py → to_master_resume_json() so experience gets populated.

Proposed solution

Acceptance

• A real PDF resume produces a non-empty experience array
• Bullets keep their order
• agent/run.py boots clean and the tailor module sees the bullets

Notes

• Don't try LLM-based parsing yet — keep it deterministic so it works without Ollama
• Test against 3-5 different real resumes (different formats) before considering done

Alternatives you considered

No response

Are you up for contributing this?

• Yes, I'd like to send a PR
• Yes, with some guidance
• No, just suggesting