ReCrew – tech recruitment CRM
Category: Client SaaS (single-tenant) Status: Delivered; last push May 2026 Time window: built early 2026 Live URL: private (client-hosted)
Purpose
Tech recruitment CRM with a CV-to-structured-data pipeline and hybrid semantic-plus-rules candidate scoring.
Who it is for
A Norwegian tech recruitment agency – recruiters and account managers matching candidates to open roles.
Stack
- Frontend: React SPA
- Backend: Express API (Node.js)
- LLM / AI: Anthropic Claude API for CV parsing and semantic scoring
- Data: structured candidate + role storage
- Deployment: Trigger.dev orchestration + client-hosted app
My contribution
Worked directly with the agency owner on the brief, shaped the recruitment workflow and the inputs the system needed to produce good matches, and carried the application from architecture and implementation through delivery and iteration.
Notable capabilities
- CV → structured-data pipeline (PDF / DOCX in, canonical JSON out)
- Hybrid scoring: deterministic rule matches (must-haves, seniority bounds) plus semantic scoring for role fit
- Single-tenant architecture – one agency, one deployment
- Norwegian + English CV handling
Outcome
Delivered as a working single-tenant CRM. Documented with WIKI.md and TODO.md per the Stratas product-doc standard.
Evidence
- Repo: private
- Docs:
WIKI.md,TODO.mdin-repo
Skills demonstrated
react-spa, express-api, nodejs, claude-api, cv-parsing, hybrid-scoring-deterministic-plus-semantic, bilingual-cv-handling, single-tenant-architecture
Want to discuss the decisions behind it?
Over fifteen years of consulting and growth, and hands-on AI delivery from discovery to daily operations. I can share my CV and talk through how I connect business outcomes with systems that run.
Available for a chat. In person in Stavanger, or remotely worldwide.