How it works
You're more than your resume.
mtmr ranks developers by the work they've actually shipped. Connect GitHub, link the projects you're proud of, and let your code make the case for you.
The premise
A resume is a one-page summary someone wrote about themselves. It tells you what they want you to know. We think hiring should run on the real thing — the commits they've authored, the code they've shipped, and the projects they can point to and say "I built that."
So we look at the work. Then we rank every applicant against every job by how well their actual evidence fits, and get out of the way.
For developers
From signup to a ranked application, here's the path.
Sign in with GitHub
Sign in through GitHub's standard OAuth flow — no separate password to manage. From there you decide what we get to look at.
Pick the repos that represent you
You choose exactly which repos go into the analysis — public or private, your call. Curate the projects that show what you can really do and where you want your career to go. The work you put forward is the work we assess.
We analyse the work, then drop the code
Your selected repos are pulled into a temporary analysis sandbox where we extract languages, frameworks, commit patterns, and code quality. Once that's done, the source is purged. We never store your code — public or private. Only the derived skill data and quality assessments stay in your profile.
Round out your profile
Upload a resume, drop in your portfolio URL, paste a short intro video, and set your location, timezone and rate. The resume gets parsed for years of experience and industry context so jobs can match against it.
Apply with evidence, not essays
For each required skill on a job, link a repo or PR that proves it. No cover letter, no "why do you want to work here" prompts. Your work does the talking.
Get ranked, not screened
Hirers see the strongest matches first. If your evidence fits the role and your code holds up, you go to the top of the list — regardless of pedigree.
For hirers
Post a job, get applicants ranked, message the top of the list.
Post a job in minutes
Title, description, salary, location and timezone. Then split your skills into must-haves and nice-to-haves. That split drives the entire ranking.
Applicants arrive with proof
Every application includes per-skill evidence URLs — a repo, a PR, a deployed project. Plus resume, portfolio and video. No chasing for missing context.
See applicants ranked automatically
Each applicant gets a score from 0–100. Match badges, must-have vs nice-to-have coverage counts, proficiency tiers and resume fit are all visible at a glance.
Vote on evidence to sharpen ranking
Upvote a strong repo, downvote weak proof. Your votes feed back into the quality score so the ranking gets sharper as you review.
Message the top of the list
Open a thread with anyone in the ranked list. Skip the bottom 80% of the pile and spend your time on people whose work already fits.
What goes into the match score
Every applicant is ranked against your job with a single match score. It blends signals across the work, the resume, and ongoing feedback from hirers — so a strong fit on the things that matter most rises to the top.
Skill coverage
How much of your skill list the applicant actually shows evidence for, with must-haves prioritised over nice-to-haves.
Evidence quality
Proficiency level, where the skill sits in their tech profile (primary, secondary, or exposure), and the volume of real code behind it.
Code quality
Per-repo quality grades from our static + LLM analysis pipeline — correctness, readability, maintainability, error handling, and codebase fit.
Hirer feedback
Upvotes and downvotes on evidence URLs across the platform feed back into how strongly any given repo or link counts.
Resume fit
Years of experience against what the role requires, plus industry overlap (exact or adjacent).
Adjacent aptitude
Strong skills the applicant has that aren't on your list but tend to come up in the same kind of work.
What we don't ask for
The performative bits of traditional hiring? We cut them. They take a long time to write, take a long time to read, and tell you very little.
- Cover letters
- Trick questionnaires
- "Tell me about yourself" essays
- Inflated job summaries you have to decode
A note on your code
You pick which repos go into the analysis, and we never keep the code itself. Source is processed in a temporary sandbox and purged once the assessment runs — we retain only the derived skill data and quality scores. That holds whether the repo is public or private. See the Privacy Policy for the full picture.
Ready to try it?
Post a job and see ranked applicants, or build a developer profile and apply with proof.