Skip to content
Start a project

AI agents · Hiring

AI hiring that ranks every CV against the job and grades the test.

Hiring teams get more CVs than they can read, and keyword filters miss people who describe the same skills in other words. We build hiring tools that rank by meaning and grade tests automatically. Our proof is SkillMatch, our own platform: more than 1,500 candidates have taken over 4,000 assessments (platform records, Oct 2026).

candidates assessed
1,500+
assessments taken
4,000+
question types graded automatically
5
skillmatch.tech
Live

What it does

  • Job descriptions enriched by AI, and applicants ranked by how well each CV fits the job.
  • CV upload that fills in the candidate profile.
  • An assessment builder with AI generated questions, automatic grading and PDF result reports.
  • Interview scheduling with Google Calendar, team roles and permissions.
  • An SDK for a company's own careers page.
SkillMatch home page: post a job, rank applicants by an AI résumé match score and send proctored assessments that are graded automatically.
The SkillMatch home page. (Live site, Oct 2026)
1 of 2

How ranking and grading work

  • CVs and job descriptions become embeddings in MongoDB Atlas Vector Search, and a model scores each CV against the job.
  • Multiple choice, coding, React, open ended and short answer questions are graded automatically.
  • Coding answers run in a self hosted Judge0 sandbox, apart from the main servers.
  • Webcam snapshots, tab switches and fullscreen exits are recorded. Each company chooses what is recorded for each assessment.

How it fits together

  1. People

    • CompaniesPost jobs and review ranked applicants
    • CandidatesUpload a CV and take proctored tests
  2. Apps

    • Next.js web appJobs, CV builder and assessments
    • SDKJobs on companies' career pages
    • Admin console
  3. Services

    • NestJS APIRoles, plans, billing and Google Calendar interviews
    • Redis and BullMQBackground jobs
    • Judge0Runs candidates' code
  4. AI and data

    • OpenAI and Gemini modelsCV ranking, job descriptions and questions
    • MongoDB AtlasWith vector search
How SkillMatch ranks applicants and grades tests

What the build took

Team
24 engineers plus PM, QA and design
Timeline
Sep 2023 to today; the first version shipped as Pre-screen
Stack
Next.js, NestJS, MongoDB Atlas with vector search, Redis and BullMQ, OpenAI and Gemini models, Judge0
Status
Live

How we deliver

  1. Scope the use case

    We pick one workflow, list its data sources and write down what a correct answer looks like.

    Output Use-case brief and a first test set

  2. Design the agent and its guardrails

    We define the tools, access rules and model, and mark the steps where a person reviews.

    Output Architecture, tool list and permission map

  3. Build in milestones

    A working agent comes early. Each milestone adds tools or data you can try on real cases.

    Output A running agent at every milestone

  4. Evaluate and run in production

    We measure accuracy on the test set, then launch with logging and review queues in place.

    Output Evaluation report, runbook and a live agent

Shipped work

Questions about AI hiring

Short answers to the questions we hear most often.

Which question types can be graded automatically?

Multiple choice, coding, React, open ended and short answer.

How is cheating handled?

Webcam snapshots, tab switches and fullscreen exits are recorded, so reviewers can see them. Each company sets what is recorded for each assessment.

Can it run on our own careers page?

Yes. SkillMatch has an SDK for companies' career pages.

Is SkillMatch live?

Yes. Companies post jobs and candidates take tests on skillmatch.tech (opens in a new tab).

Have an agent, protocol or product to ship?

Send a short brief and we'll reply with times for a 30-minute scoping call.