Six steps to a defensible shortlist
Open a session. Pick the branch. Weight what matters. Let the engine score every candidate against the role, software and core engineering alike, and rank them with reasons you can stand behind.
Open a session
A session is one recruitment drive: "TCS NQT · June 2026", "L&T Core Hiring · Batch 2027". Each is a private, isolated workspace with its own candidates, weights, and rankings. Sessions are organised under branches like CSE, ECE, Mechanical, and Civil, and each branch carries its own scoring profile, so a software drive and a core-engineering drive are judged on the evidence that fits them.
Set your weights
Not every company values the same things, and neither does every branch. Tune how much each signal contributes: academics, hands-on evidence, contest performance, résumé depth. A product firm might lean on coding evidence; a core-engineering drive might lean on GATE, certifications, and CGPA.
Add the candidates
Publish a shareable Form Builder link that students fill out themselves. It pre-loads all required fields and exports directly as an engine-ready spreadsheet. Or upload a prepared Excel sheet, or let the AI Excel Structurer clean any messy one you already have.
Let the engine score
CandidateRank gathers live evidence from public builder profiles (GitHub, Kaggle and the like), reads performance on problem-solving platforms (LeetCode, Codeforces, Hackster, Rosalind), takes GATE scores for core branches, and parses every résumé with AI, scoring each candidate against the role rather than a keyword list. Watch progress live, per student.
Query your batch with AI
Ask the AI Advisor anything in plain English: "Who has the strongest DevOps background?", "Find candidates who interned at top companies". It searches every scored candidate and surfaces ranked matches with the evidence behind each one.
Rank, shortlist, and share
An ordered leaderboard where every position is explained. Tag candidates as Shortlisted, Rejected, or Scheduled as you move through rounds. Compare any two side by side. Export a branded PDF or share a live public link with recruiters: one URL, always current.
Build your talent pool before companies ask
Hivemind lets students submit their own profiles through a public portal link. No spreadsheets, no chasing emails. Your cell approves submissions, and CandidateRank builds a living, ranked talent pool across all batches automatically.
Admin opens the Collection Portal
Generate a shareable portal link from the Hivemind tab. One link covers your entire institute, with no per-session setup needed.
Students self-submit their profile
Students open the link and submit their profile: coding and portfolio handles, résumé, and CGPA. They can verify ownership of their developer profiles via OAuth so you know the data is authentic.
Admin reviews and approves
Submissions land in the Review Queue. You approve, request changes, or reject each one, keeping full control over what enters the talent pool.
A ranked talent pool builds automatically
Approved profiles are scored and ranked by domain strength: Full Stack, Data Science, ML, DevOps, and more. When a company asks for candidates, your shortlist is already ready.
What we actually measure
The exact mix adapts to each branch: software drives read code and contests, while core branches lean on GATE and hands-on evidence. Only verifiable data counts toward a rank.
Builder Portfolios
Public proof of work from platforms like GitHub and Kaggle: code, datasets, models, hardware builds. The truest window into how a candidate actually builds.
Problem-Solving Arenas
Ratings, contest history and challenge performance from platforms like LeetCode, Codeforces, Hackster and Rosalind, revealing depth no exam paper can surface.
National Benchmarks
For core branches like Mechanical, Civil, Electrical, Chemical, and Biotech, a GATE score anchors the ranking in a nationally recognised, verifiable exam.
Résumé & Evidence
AI-parsed résumés enriched with branch-specific evidence: projects, internships, certifications, competitions, and research. Initiative beyond class gets rewarded.
Academic CGPA
One signal among many: not dismissed, but proportionally weighted so it enriches rather than dominates the rank.
Tune the weights per drive
Every recruiting company is different, and so is every branch. Each discipline starts from its own default mix, and you adjust how much each signal counts. The ranking recomputes instantly, fully auditable and defensible to the last decimal.
Bring your own pipeline
Publish a Form Builder link that students fill out, upload an Excel sheet, or let the AI Excel Structurer tidy any messy one. The engine works with the data your placement cell already has.
Role-aware scoring + AI search
Live evidence from builder and contest platforms, GATE scores for core branches, and AI-parsed résumés, tuned to the work rather than the wording. Then query the scored batch in plain English with the AI Advisor: "Who has the strongest backend experience?" It finds them.
Explainable ranks
Every score carries its reasoning. Tag candidates through pipeline stages, compare any two side by side, export as PDF, or share a live leaderboard link with recruiters.
Built for the whole training and placement cell
CandidateRank covers the full workflow, from collecting candidate data to sharing results with companies, with a role-based team layer and audit log underneath.
Multi-role team access
Invite coordinators as Admins, Sub-admins, Members, Viewers, or external Recruiters. Each role sees only what they need: no shared logins, no accidental deletes.
Branches per department
Group sessions by department or year: CSE, ECE, Mechanical, Civil. Each branch is isolated and scored on its own profile. Sub-admins manage their own; the Admin, usually the training and placement officer (TPO), sees everything across the cell.
Placement toolkit
Form Builder for data collection, AI Excel Structurer for messy sheets, Word-to-PDF converter, and Files-to-ZIP bundler, all in one place with no third-party tools needed.
A record that defends you
Pressure on a placement cell is personal: a call about someone’s nephew, a HOD with a favourite. Every decision here is timestamped against the ranking in force at the time, and an override can carry your own reason, so the answer is procedural instead of personal.
From five hundred applications to a shortlist you can defend.
The point isn’t to shortlist faster. It’s to shortlist better, and to be able to say exactly why. Cells using CandidateRank spend less time sorting spreadsheets and more time talking to the right students.