About
Crosshire was built to fix two broken sides of the same problem.
// WHY THIS EXISTS
Talented candidates spend weeks applying to roles they were never going to get. Recruiters drown in CVs that don’t match the role they posted. Both sides lose.
The job market doesn’t have an information problem. It has a signal problem.
Most “AI job tools” make it worse — helping people apply to more roles, faster, with even less context. That’s volume, not signal.
Crosshire takes the opposite approach: read deeply, score honestly, cite the evidence, and tell the user the truth. For both sides.
// WHAT WE BELIEVE
A score is only useful if you can see how it was made. That's why every Crosshire score is broken down with cited evidence and explicit gaps.
AI tools should never hallucinate skills. Every claim in your Summary is matched back to your actual CV text. Anything we couldn't verify is flagged.
Honesty about gaps is more valuable than flattery. We tell you what will hurt in screening — and what to do about it — before you waste an hour on a cover letter.
Applying smarter beats applying harder. Five thoughtful applications beat fifty thoughtless ones. Crosshire is built for the candidate who'd rather know the truth than play the volume game.
// WHO CROSSHIRE IS FOR
Crosshire is currently tuned for data, software, and AI roles across DACH and UK markets. The taxonomy, scoring calibration, and skill ontology are built for these roles specifically.
If you’re a candidate or recruiter in adjacent fields, Crosshire still works — just with less precision. We’ll expand into more verticals as we grow.
// HOW WE BUILD
Crosshire is built in Europe, hosted in the EU, and designed to be GDPR-native by default — not as an afterthought.
We use Anthropic Claude and Google Gemini for parsing and reasoning, with EU-region data processing agreements. Your CV is encrypted at rest, deletable on request, and never sold or shared.
We’re a small, focused team building this carefully. Crosshire grows when it earns the trust to grow.
// THE TEAM
Every line of code in Crosshire was written by AI agents. The human writes the instructions, makes the product calls, and runs the final manual checks. Haiku and Gemini handle the end-to-end test runs. Each agent plays a specific role; nothing ships without human review.
Darshan Singh
Founder / Data Engineer
Writes the instructions, reviews every change, makes product decisions, and runs the final manual checks. Does not write production code.
SR
Inspiration
MS
Inspiration
ASM
Inspiration
Claude Opus
Lead Architect
Decomposes complex changes, plans architecture, decides what to build and in what order.
Claude Sonnet
Main Implementer
Writes most of the production code. Solid at well-specified backend and frontend work.
Claude Haiku
Pipeline + E2E Testing
Runs the LLM parsing and scoring pipeline in production, and drives the end-to-end test sweeps against the live app.
Claude Design
Web Design
Visual polish, layout, theme tokens, and the editorial cream aesthetic of the Summary view.
Gemini Pro
Deep Tech Reviews
Second-opinion reviewer for thorny architectural calls and migration safety.
Gemini Flash
E2E Testing + Light Reviews
Drives end-to-end test runs against the live app, plus quick reviews, log triage, and reproduction help when something is misbehaving.
Transparency note: the code is written entirely by AI agents. A human reads and approves every change before it ships, writes the prompts that the agents work from, and runs the final manual checks. The credit roll above is the actual team.
// CONTACT
For recruiter access, partnership inquiries, or feedback from your job search — please reach out. We read every message.
Worth applying?
Sign in →Crosshire is an independent project. Not affiliated with any recruiter, job board, hiring platform, or AI provider. Claude and Claude Code are products of Anthropic. Gemini is a product of Google. LinkedIn® is a registered trademark of LinkedIn Corporation.