AI resume screening · built for recruiters

Turn 200 resumes into a ranked shortlist before lunch.

Upload resumes and a job description. ShortlistTable extracts candidate details, scores fit, flags missing requirements, and shows evidence from every resume.

Join the waitlistSee how it works
No credit card·CSV export included·You stay in control of every decision
No more copy-paste/No more resume pile/No black-box scoring
All jobs/Senior Backend Engineer
Bramwell Capital · Toronto, Hybrid
Senior Backend Engineer
5+ yrs · Python · Kafka · Distributed systems
AskingKafka experienceFintech backgroundLed a team?
Drop 200 resumes here
PDF · DOCX · TXT
Built for recruiting teams at
KESTREL & COMAYFAIR TALENTNorthboundOAKWOODBRIDGEPATHStellate
The pile problem

The best candidates are hiding in a pile of PDFs.

Every new role creates the same mess: resumes from LinkedIn, Indeed, email, referrals, and your ATS. Different formats. Missing fields. Too many tabs. Too much copy-paste.

01Manual drag

Screening becomes data entry

Every source lands differently, so your team burns time opening files, copying fields, and rebuilding the same candidate table by hand.

LinkedIn, Indeed, email, ATSMissing location, years, skills
2.5 hrsto copy-paste 200 resumes
02Slow signal

Strong candidates wait too long

While the pile is still being sorted, the best applicants are invisible to the team and already taking calls somewhere else.

No ranked viewNo immediate call list
48 hrsbefore strong applicants accept elsewhere
03Scattered tools

Parsing and prompts don't make a workflow

ATS parsers extract generic fields. ChatGPT handles one resume at a time. Neither gives you custom screening columns, citations, exports, and a review queue.

12 fixed fields vs role criteriaNo batch review or source trail
1 → 200different problem entirely
New · built for the age of AI resumes

Anyone can write a perfect resume now. Few can back it up.

AI lets every candidate generate a keyword-perfect resume tailored to your exact posting. Keyword matching can't tell them apart — so we don't keyword-match. We read for substance: concrete outcomes, real systems, measurable scale. The fluff gets flagged, with the evidence.

AI-tailored resumeLooks qualified
Experienced with Python, Kafka, and distributed systems
Proven track record building scalable, production-grade platforms
Results-driven engineer passionate about clean architecture
Led migration to a modern microservices stack
Passes a keyword screen. Tells you nothing.
What we surfaceActually qualified
Built Kafka pipeline processing 4M events/day across 40 topics
Cut p99 checkout latency 820ms → 90ms on a 12-service system
“clean architecture” — no specifics, no system, no outcome
“led migration” — no scale, no role, no result given
Real depth ranked up. Fluff flagged — with the quote.

We don't guess whether a resume was written by AI — an unreliable, accusatory game. We measure something real: whether the experience is substantiated.

How it works

Five steps from inbox to call list.

No prompt engineering. No workflow builder. Recruiter inputs only — and the table does the rest.

01

Upload resumes

Drag in PDFs, DOCX files, or text. From job boards, LinkedIn, email, your ATS.

02

Add the job description

Paste the role requirements once. Used to score every candidate.

03

Tell the table what to check

Add columns like 'Kafka experience', 'Toronto-based', 'Worked in fintech'.

04

Review the shortlist

See match scores, missing requirements, red flags, and source evidence per cell.

05

Export and contact

Send to CSV or XLSX. Get on the phone.

Product proof

One workflow: screen, verify, shortlist.

Instead of spreading the product across separate demos, this is the core loop: define what matters, check every candidate, and act on the ranked list.

01 / Custom columns

Ask the table role-specific questions.

Add screening columns for must-have skills, domain context, location, seniority, sponsorship, or whatever this role actually needs.

Plain-English criteriaReusable role templatesBatch re-read when columns change
All jobs/Senior Backend Engineer
Bramwell Capital · Toronto, Hybrid
Senior Backend Engineer
Add screening columns for this exact role.
AskingKafka experienceFintech backgroundLed a team?Needs sponsorship?Built high-scale systems?
CandidateMatchYearsLocationKafka exp.Status
SO
Sasha Okafor
Backend Engineer
954ySan Francisco, CAUnclearCall first
LH
Lila Hernandez
Backend Developer
955yToronto, ONNoCall first
CD
Camille Dubois
Senior Backend Engineer
955yMontréal, QCNoCall first
AR
Aisha Rahman
Backend Developer
956yBengaluru, INNoCall first
AC
Alice Chen
Senior Backend Engineer
928yToronto, ONYesCall first
ROI · time math

How much does resume screening cost your team?

Punch in your numbers. If a recruiter spends 45 seconds copying basic details from each resume, 200 resumes is 2.5 hours before real screening even starts.

Inputs · drag to adjust

150
8
45 sec
$55/hr

Result · your team's math

Resumes per month
1,200150 × 8 roles
Manual hours per month
15 hrsat 45 sec/resume
Manual cost per month
$825at $55/hr loaded
Time saved with ShortlistTable
13 hrs≈ 85% of manual screening time
Cost saved per month
$701vs ShortlistTable Team plan at $199/mo
Use cases

Built for the way your team hires.

Recruiting workflows aren't one-size-fits-all. ShortlistTable bends to the shape of yours.

01 / Agencies

Staffing agencies

Review high-volume applicant batches and submit strong candidates faster than the agency down the street.

high-volume batchesclient-ready exportsspeed-to-submit
4xfaster shortlist handoff
01Bulk import
02Client-specific columns
03Shareable shortlist
Objections, fairly handled

The questions every recruiter asks first.

ATS parsing extracts basic fields like name, email, phone, and last role. ShortlistTable helps you screen against a specific job description, add custom criteria your client cares about, rank candidates, and verify every answer with source evidence.
ChatGPT works for a few resumes. ShortlistTable is built for batches: tables, saved screening columns, review queues, confidence scores, source evidence, exports, and a workflow your team can repeat across roles.
Every answer includes a confidence score and source evidence — the exact resume sentence behind it. Low-confidence cells are flagged for your review first, and you can edit or rerun any cell.
Candidate documents stay private to your workspace. Access is workspace-based, you control retention, and your data is never used to train public models. SOC 2 is on the roadmap.
Yes — that's the point. Add a screening column for the exact requirements of each role. Save templates for the searches you run often.
PDF, DOCX, and plain text. Drag a folder or upload individually. We extract structured fields and keep the original file for reference.
One more thing

Be first to screen for substance, not keywords.

We're onboarding recruiters in small batches. Join the waitlist and you'll be in the first group to get early access.

Join the waitlist

Your shortlist is ready

200resumes processed
32strong matches
41need review
19missing must-haves
6low confidence