AI question generation
Generate per-candidate question sets from a role blueprint and difficulty curve. Every item is reviewable and editable before it goes live.
Products — Assessment & hiring
Run the drive. Not the spreadsheet.
The assessment and hiring-drive engine behind our recruitment work — AI-assisted question generation, a real coding environment, defensible proctoring, interview workflow and a report that arrives while the decision still matters.
0Drives run
0Partner institutions
01 — The problem
What actually goes wrong on drive day
Eight hundred candidates arrive. Four rooms, one of which has no working projector. The assessment platform was tested on the office network, not the campus network, and it starts timing out at candidate two hundred. Someone begins writing names on paper.
The panel interviews until seven in the evening against a rubric that was agreed verbally that morning and interpreted differently by each interviewer. Shortlisting happens in a corridor conversation.
Three weeks later a spreadsheet arrives. By then two of the strongest candidates have accepted elsewhere, and nobody can reconstruct why candidate 412 was rejected.
None of this is a hiring problem. It is an operations problem, and it is entirely solvable with software plus people who have run drives before.
02 — The solution
Drive is the operations layer
One system holds registration, eligibility, assessment, proctoring, panel scheduling, rubric scoring and reporting. Because everything is in one place, the report is a query rather than a reconstruction — and it is available the moment the drive closes.
The assessment layer runs a real code execution environment against hidden test cases, so a coding round measures working code rather than syntax recall. Question sets are generated per candidate from a blueprint, which makes leaked papers largely worthless.
Proctoring produces evidence, not verdicts. The system flags what it observed; a human reviews every flag before it affects a candidate. Candidates are told in advance exactly what is monitored.
03 — Capabilities
What's in it.
9 capabilities, each of which exists because something specific went wrong without it.
Coding platform
Real execution across 14 languages against hidden test cases, with time and memory limits and partial scoring.
Remote proctoring
Tab-switch, face-absence and multi-face detection with a session recording. Flags are evidence for human review, never automatic disqualification.
Drive orchestration
Registration, eligibility rules, slot allocation, venue capacity and check-in — the logistics, handled.
Interview workflow
Panel scheduling, structured rubrics, per-round scoring and written recommendations captured in the room rather than afterwards.
Mock drives
Full-fidelity practice runs for students on the real interface, so drive day is not their first encounter with it.
Reports in 48 hours
Per-candidate evidence, round-by-round scoring, funnel analytics and a shortlist you can act on immediately.
Offline resilience
The assessment client buffers locally and syncs on reconnect. A network drop does not cost a candidate their attempt.
Integrations
Push shortlists to your ATS, pull cohorts from LARE LMS, and wire the whole thing through the REST API.
04 — Dashboards
Three audiences, three views.
Each role sees the question it actually needs answered — not the same dashboard with rows hidden.
Company dashboard
Create a drive, watch it run, decide from evidence.
- Drive creation from role templates
- Live progress: registered, checked in, in progress, completed
- Candidate profiles with code, scores and proctoring evidence
- Rubric-based panel scoring and consolidated recommendations
- Funnel analytics and per-institution comparison
- Shortlist export and ATS push
Student dashboard
Know what's coming, practise on the real thing, see where you stand.
- Upcoming drives with eligibility status
- Mock drives on the identical interface
- Attempt history with score breakdown
- Practice sets targeted at weak areas
- Interview schedule and result notifications
Institution dashboard
The view your T&P cell has been building manually every placement season.
- All drives on campus, one calendar
- Student participation and conversion tracking
- Branch-wise performance against recruiter expectations
- Recruiter feedback aggregated across drives
- Season reporting for management
05 — How it runs
End to end.
The same sequence every time, which is what makes the reporting a query rather than an archaeology project.
- 01Company
Define the role
Role blueprint, eligibility rules and difficulty curve. The assessment is generated from this, not chosen from a catalogue.
- 02Institution
Publish to campus
The drive appears on the placement calendar; eligible students are notified and register in-platform.
- 03Student
Mock drive
Candidates practise on the exact interface, so drive day is about the questions rather than the software.
- 04Student
Assessment
Per-candidate question set, real code execution, proctored session, local buffering against network loss.
- 05Company
Review & shortlist
Scores, submitted code and proctoring evidence in one view. Flags reviewed by a human before anyone is cut.
- 06Company
Interview rounds
Panels scheduled automatically, scoring against the pre-agreed rubric, recommendations written in the room.
- 07All
Report in 48 hours
Company gets the shortlist and funnel. Institution gets participation and conversion. Students get their results.
06 — Why this one
What makes it different.
Built by people who run drives
Every feature exists because something went wrong on a real drive day. The offline buffering is there because a campus switch failed at candidate 200.
Proctoring that survives a challenge
Flags are reviewed by humans and candidates are told in advance what is monitored. A drive that disqualifies people by algorithm will eventually be challenged, and should be.
Generated question sets
Per-candidate generation from a blueprint means a leaked paper is worth very little, and difficulty stays comparable across candidates.
Forty-eight hours, not three weeks
Because the data was structured during the drive rather than reconstructed after it, the report is a query. Speed here directly changes offer-acceptance rates.
07 — In production
Where it stands.
0Drives run
0Partner institutions
0hReport turnaround
0Languages supported
08 — Pricing
What it costs.
Published figures are pending finalisation — request pricing and you'll get a real number against your actual cohort size, not a range.
Per drive
One-off campus or walk-in drives.
₹—per drive
- Up to 500 candidates
- Assessment & coding platform
- Standard proctoring
- Interview workflow
- 48-hour report
- Email support
Annual
Most commonOrganisations hiring continuously through the year.
₹—per year
- Unlimited drives
- Everything in Per drive
- AI question generation
- Custom rubrics & role templates
- ATS integration
- Analytics across drives
- Named success manager
Institution
Colleges running their own placement season on Drive.
Customper campus
- Placement-cell dashboard
- Unlimited mock drives for students
- Recruiter onboarding
- LARE LMS integration
- Season reporting
- On-site support on drive day
See LARE Drive on your own data.
Thirty minutes in a working environment with your scenario loaded. If it's a bad fit we'll tell you in the call rather than three emails later.