The AI hiring panel
that asks back.
For founders hiring outside their expertise. Lucida reads the JD, asks what a senior recruiter would ask, then scores every candidate with cited evidence. You do the human call.
Strong backend + ops. Verify observability + mentorship depth in the live round.
The whole loop, on autoplay.
Paste a JD. See what happens next. No signup.
JD says "5+ years" AND "IC role, no direct reports." Is a 12-year staff-eng in scope?
"Postgres-deep" — operational (scale, on-call) or design (schema, planning)?
“Built our on-call rotation fairness framework — stopped the burnout spiral.”
Strong ops + scale signals, zero interview answers — use this call to verify reasoning.
This runs on autoplay. Start free to run it against your own JD.
Three hiring moments Lucida is built for.
You can write the JD. You cannot interview a marketer. Lucida asks the four questions you would not have thought to ask — before the rubric.
Forty-seven résumés stacked in your LinkedIn inbox. Paste them all at once. Six excellent surface in nine minutes, every score backed by quotes.
Every applicant — the ones you reject too — gets a real feedback report. Some come back with sharper answers. Some refer a friend.
What you're paying right now to hire badly.
The typical Indian SMB hiring stack — three tools, none of which talk to each other, none of which actually score. Numbers below are sticker prices from the vendors' own sites.
~92% lower. One tool. One bill. One workflow.
Hiring outside your expertise is a gamble — every time.
A technical founder hiring a growth lead. A designer hiring an engineer. You know the interview went badly. You can't tell if the candidate is bad — because the résumé is a story they've rehearsed.
Burn hours per hire. Coin-flip the ambiguous ones. Trust drops each time.
They send anyone with a pulse — their incentive is volume, not signal. You still do the call.
Convince yourself a generalist can cover it. Until the growth miss, design bug, or pipeline gap makes it obvious.
Illustrative — the point is not the number, it's that the cost of hiring badly compounds silently.
Lucida is the rubric-backed panel a 500-person company runs — for founders who wear all the hats. AI that asks the right questions before it scores, not after.
Not a claim — the actual math for a founder running 3 roles a year.
Paste a JD. Do the human call. Ignore everything in between.
Your whole loop, in four steps. Scroll to walk it.
Paste the JD. Lucida asks back.
Before drafting anything, Lucida reads your JD and surfaces the three-to-five questions a senior recruiter would ask. Thirty seconds of answers — then the five-dimension rubric and ten interview questions are generated against your clarified spec.
Candidates take the interview.
Your careers page goes live at lucida.team/c/yourcompany. Or skip the careers page — paste the 60 résumés stacked in your LinkedIn inbox in one go. Lucida scores every one.
Every score cites the résumé.
Each candidate is scored per dimension, with direct quotes from their résumé and answers as evidence. If the quote isn't there, the score doesn't go up.
Shortlist → they book. You show up prepared.
One click shortlists a candidate. Lucida emails them your Cal.com link and generates a two-minute briefing for you — with the exact concerns to verify in the human call.
The JD says "5+ years" but also "IC role, no direct reports." Is a 12-year staff-eng in scope, or does this cap at a certain seniority?
You mention Postgres-deep. Is that operational (scale, migration, on-call) or design (schema, query planning)? Different candidates.
Mentorship depth — nice-to-have or must-have? If must-have, weight goes to 25% and we change the questions.
"Built our on-call rotation fairness framework — stopped the burnout spiral"
Strong ops + scale signals, zero interview answers — use this call to verify the reasoning behind the résumé.
- 01Walk me through the 500M-row Postgres migration — what did you reject?
- 02Observability isn't on your résumé — how did you know the service was healthy?
- 03One concrete mentoring moment — not the outcome, the moment.
Paste the JD. Lucida asks back.
Lucida reads your JD and surfaces the three-to-five questions a senior recruiter would ask. Thirty seconds of answers — then the rubric and questions are generated.
Candidates take the interview.
Careers page goes live at lucida.team/c/yourcompany. Or paste the 60 résumés from your LinkedIn inbox in one go.
Every score cites the résumé.
Each candidate scored per dimension, with direct quotes as evidence. If the quote isn't there, the score doesn't go up.
Shortlist → they book. You show up prepared.
One click shortlists. Lucida emails your Cal.com link and generates a two-minute briefing for you.
Three things Lucida does that nothing else in the category does.
We looked at Keka, HireQuotient, Zoho Recruit, Ashby, and ChatGPT. Everyone scores. Only one product does these three.
The JD says "5+ years" AND "IC role no direct reports." Is a 12-year staff-eng in scope?
"Postgres-deep" — operational (scale, on-call) or design (schema, planning)? Different candidates.
Mentorship — nice-to-have or must-have? Weight moves from 12 to 25 if must.
Reads your JD, then asks the questions a senior recruiter would.
Everyone else scores against whatever JD you uploaded — ambiguities included. Lucida catches those ambiguities first, asks you to resolve them in 30 seconds, then builds the rubric against a clarified spec. The clarification IS the moat.
Every score quotes the exact line it came from.
Everyone else gives you a black-box number. Lucida gives you a number + the quotes from the résumé and interview that produced it. If the quote isn’t there, the score doesn’t go up. No hallucinated strengths, no invented evidence — because we made the citation the first-class primitive, not an afterthought.
Strong signal of on-call ownership + team-level framework to prevent burnout.
- “Built our on-call rotation fairness framework — stopped the burnout spiral”
- “Postmortem facilitator for a customer-facing outage caused by a noisy-neighbor tenant”
- “Led a Postgres partitioning migration on a 500M-row orders table. Zero downtime”
Strong operational ownership — on-call framework + 500M-row Postgres migration are exactly what the role needs.
Distributed-systems judgment came across thinner than the role needs. No message-queue failure-mode examples.
The most thoughtful rejection they have ever received.
Recruiters cannot afford this — 20 minutes per rejected applicant times ₹2L cost-per-hire kills the math. AI can. Every applicant, even the ones you reject, gets a private URL within 24 hours: strengths, gaps, cited evidence, what to do next. Each report signs off “made with Lucida” — the next three candidates apply because of the first.
Most tools forget you between hires. Lucida remembers.
Every job sharpens your rubric. Every score teaches us which dimensions predicted the strong hires. Six months in, Lucida reads a new JD the way a recruiter who's worked with you for years would.
Hire #2 starts at hire #1's rubric.
One click clones the last role's weights + questions + knockouts. Edit the title, ship in 30 seconds. Zero extra cost.
See how other founders are doing.
Aggregate-only. "8 other founders on Lucida are hiring senior React engineers in Bangalore. Average score 78. 14% hit excellent." You know if your threshold is sane.
90 days later, we ask.
One email: "Did you hire them? How's it going?" Your reply tells us whether the score predicted the hire — and tunes the next rubric.
You become the kind of founder who has institutional hiring memory — not just a stack of one-off interviews you'll forget by next quarter.
Lucida also tells you where to post the role.
Most hiring tools score what walks in. We help you fill the door too — and measure what each channel produces. Post-everywhere-and-hope becomes per-channel ROI.
Smart channel recommender.
Lucida picks 5 channels for the role + region + seniority — free (LinkedIn, subreddits, HN), cheap (Wellfound, Demand Curve), paid (Naukri, Instahyre, iimjobs). Each with a one-line reason.
Clicks → apps → scored → hired.
"LinkedIn: 32 clicks → 8 apps → 3 scored ≥80 → 1 hired." By hire #3, you stop posting where the funnel leaks.
Embed the feed on your site.
Indeed XML + JSON feed + embeddable iframe widget. One careers list, three surfaces, one workflow.
One tool, end-to-end: where to post, how to score, who to call. No tab-switching between three vendors. Every applicant flows through the same workflow, every channel reports back into the same ROI table.
Designed for how Indian SMBs actually hire — not localised on the way out.
We didn't translate a US tool. Indian hiring has its own gravity — WhatsApp not email, ₹ LPA not USD, Diwali dead zones, tier-2/3 candidates who deserve a fair score.
LPA or ₹/month TPM · CTC or in-hand · auto-converted. No more "5 LPA means how much in hand?" friction.
Apply-confirmation + Round-2 nudges over WhatsApp, not lost-in-spam Gmail. Tenant opt-in.
Hindi-English mix on the apply form. Future: Tamil, Telugu, Marathi, Bengali. Candidate keeps their voice.
Diwali, Holi, Eid, Pongal, Onam — your dashboard surfaces "expect 40% lower apply rate this week." No silent confusion.
Razorpay India · GSTIN field on the workspace · one downloadable invoice per cycle. Built for your CA, not Stripe Atlas.
Kirana-tech, BNPL/fintech, agri-fintech, D2C, gig economy, edtech-Bharat. Six starter rubrics that know your domain.
Pages under 30KB, no SPA, no Tailwind CDN in prod. Loads on a 2018 Redmi with 200ms latency. Candidates apply, not bounce.
No dosha typing. No numerology. No photo features. No marriage questions. No caste filters. Hiring is a fairness contract — we keep ours.
Full rationale on every line we will not cross is at /who-we-are-not-for. DPDPA 2023, EU AI Act Art. 26, and NYC AEDT compliance covered below.
Built by one person.
Priced like it.
No enterprise sales team to fund. No customer-success managers to fund. No quarterly seat-expansion targets to hit. The product stays small and focused — and your invoice does too.
See what a Lucida rubric looks like — for your role.
Pick a role. See the 5-dimension rubric Lucida would generate for a senior hire — weights and all. No signup, no card. The fastest way to see the shape of the output.
Three tiers. No per-seat.
Start monthly. Switch to annual when the score graph proves itself. Or pay once and keep it forever. No per-candidate fees. No dunning emails.
For the founder trying it on their next hire. Cancel anytime; upgrade to annual when the score graph proves itself.
- Unlimited jobs + candidates
- AI rubric + 10 questions per job
- Adaptive interview, rubric-backed scoring
- Round-2 briefing on every shortlist
- Bulk-paste import (60 résumés in one go)
- Candidate feedback reports — every applicant
- Distribution recommender + channel ROI
- Indeed XML feed + embeddable jobs widget
- Email alerts on every scored candidate
For the founder hiring three or more roles this year. Same full product, ₹5,000 saved vs monthly, one GST invoice for the year.
- Everything in monthly — every feature, no gates
- Save ₹5,000 (17% off)
- One GST invoice for the year
- Priority email support — founder-direct
For the founder building a team they'll keep for years. Pay once, locked pricing forever, your name on the Founding 200 list.
- Everything in annual prepay
- Locked pricing — no renewal, ever
- Founding 200 badge on your careers page
- Direct line to the founder on roadmap
- Lifetime access to every new feature we ship
All prices in INR · billed via Razorpay · GST invoice on request · no auto-renewal traps.
Honest answers to the fair ones.
How does the scoring actually work?
When you publish a job, Lucida AI reads the JD and produces a five-dimension weighted rubric and ten role-specific questions. Every candidate is scored against those dimensions with direct quotes from their résumé and interview answers as evidence. You see the reasoning, not just a number — and you can overrule the AI at any point.
What if the AI gets a candidate wrong?
The score is a first pass — it surfaces signal, it doesn't make the call. You still do the human conversation. The briefing tells you exactly what to verify in the live round, especially the things the scoring flagged as weak. Treat Lucida as a very fast, very literal junior recruiter who's read every résumé carefully.
Do you share my candidate data with anyone?
No. Your candidates' résumés and interview answers are yours. We use them only to score them against your rubric. We do not train models on your data, we don't share with third parties, and every candidate's report URL is private to the candidate and your workspace.
How is the pricing structured?
Three tiers, no per-seat: ₹2,499/month if you want to try it, ₹24,999/year if you're committing (17% off), or ₹49,999 one-time for the first 200 customers (Founding 200 LTD — locked pricing, no renewal). Every paying tier gets the same product — same rubrics, same scoring, same briefings, same distribution recommender, same channel ROI. The difference is billing cadence, the Founding 200 badge on your careers page, and your name on the founder's direct line.
Is there a free tier?
Your first ten candidates are free — enough to hire for one role and see the product in action. After that, any of the three paid tiers unlocks unlimited jobs and candidates. No per-seat fee, no per-candidate fee, no dunning emails.
Does Lucida help me find candidates, or just score them?
Both. Right after you publish, Lucida AI picks 5 channels for that specific role + region + seniority — your LinkedIn, the right Naukri category, Wellfound, HN, the niche communities — with a one-line rationale each. Free channels open a pre-filled share dialog. Paid channels deep-link with a Lucida referral code. Then we track click-through, so by hire #3 you know which channels actually produce hires for your roles, not just clicks.
Will my jobs show up on Indeed / job boards / my own site?
Yes. Every workspace gets an Indeed Publisher-format XML feed at /c/<you>/indeed-feed.xml, a lightweight JSON feed at /c/<you>/jobs.json, and an iframe-embeddable widget you can drop into your marketing site. Your careers list lives in three places without three workflows. Lucida is hiring infrastructure — not a destination you have to send candidates to.
Is this India-only?
Built India-first. Pricing in ₹, billing via Razorpay with GST, comp displayed in LPA or ₹/month with auto-convert, WhatsApp Business API ready, vernacular Hindi-English apply form, festival-aware nudges, tier-2/3 connection-speed budget. The underlying product works anywhere — but the defaults, the partner channels, and the compliance posture (DPDPA 2023) are India-native.
Who's behind this?
A solo technical founder who got tired of gut-feel hiring. Lucida is intentionally narrow — built for one type of buyer (the founder hiring outside their expertise), three pricing tiers, no per-seat math, no human in the support loop except the founder. That is the reason Lucida costs ₹2,499 a month and not ₹25,000.
Post the JD. We'll take it from there.
Sixty seconds to a rubric. Thirty seconds to a score. Your first ten candidates free. If it helps you hire, stay. If not, don't.