Contact Us
×Don't just certify what people know. Prove what they can do.
Hiring, AI interviews, adaptive learning and AI training simulations — running on one shared skill vocabulary.
Difficult conversations
Handles resistance without becoming defensive
Acknowledges, clarifies, holds the accountability line.
Argues the process or jumps straight to a fix.
Scored on transcript evidence — against a criterion you wrote.
The common thread
One skill. Four surfaces. Same vocabulary.
Pick a skill and watch it hold its shape from learning through to hire.
Adaptive learning
Training simulation
AI interview
Hiring
What's included
Four capabilities, one record of the learner
Each works on its own. Together they close the loop between learning and hiring.
Hiring & matching
Upload a job description; the platform builds a structured rubric your recruiters edit before anything scores.
- Job Post Architect — skills, tools and tasks with weights
- Hybrid engine — 70/30 rubric-to-semantic, tunable per role
- Why this candidate — reasoning attached to every shortlist
- Bulk CV import — ZIP upload, validation, no duplicates
AI interviews
A structured first round tied to the same rubric the job was built on, retained for audit.
- Question stacks — every candidate evidences the same criteria
- Proctoring — integrity rules configured per programme
- Structured reports — not a raw transcript dump
- Mock interviews — candidates rehearse before the panel
AI adaptive learning
Every gap links to a bridging course, and finishing it moves the match score in real time.
- AI coach — diagnoses why the learner is stuck first
- Adaptive pathways — skip what's known, reinforce what isn't
- Faculty guardrails — tone, scope and intervention thresholds
- Human handoff — escalate to a live tutor with context
AI training simulation
Author the situation, the AI's character and the rubric. The agent holds character and scores afterwards.
- Scenarios & personas — drawn at random each session
- Competency rubric — weight, max score, strength and risk signals
- Per-criterion feedback — evidence from the transcript
- Nothing to game — AI direction hidden from the learner
How a simulation runs
Authored once. Different every attempt.
- STEP 01
Author
Anchor to a course or competency framework, then write the rubric.
- STEP 02
Brief
The learner sees both roles, the objective and the ground rules.
- STEP 03
Draw
Scenarios drawn at random; one persona held for the whole run.
- STEP 04
Practise
The AI stays in character and gives no coaching mid-session.
- STEP 05
Score
Only at the end — per criterion, on transcript evidence.
Built for every stakeholder
Three buyers, one platform
Universities
- Configurable candidate journeys, no code
- Campaigns, NDAs and scheduling
- Cohort dashboards for placement cells
- Enrolment-to-placement reporting
Employers
- Rubrics built from your JD, edited by you
- AI first rounds with structured reports
- Bulk CV ingestion at volume
- Role-play training for interviewers
Credentialing bodies
- Surface SMEs from your holder base
- Recommend the next certification
- Show employers what the credential predicts
- White-labelled end to end
Platform & governance
Built to survive a procurement review
Dedicated instance
White-labelled, cloud-hosted, no shared tenancy.
Live in 2–3 weeks
API-first for LMS, SIS and exam platforms.
Explainable decisions
Every match and score carries its reasoning.
Accessible by default
Built to recognised accessibility standards.
Traceable operations
Dedicated log streams for end-to-end tracing.
Human in the loop
Weights are visible; a person makes the call.
See it run on your own framework.
Bring a real job description or competency framework. We'll walk the full loop with your team.
- 45 MINA working session, not a slide deck.
- SANDBOXA personalised environment for your institution.
- SCOPEStart with one capability, add the rest later.
