About Coril

Last updated May 3, 2026

Most people know their stuff. They freeze under pressure. The gap between knowing the answer and saying it clearly to someone who is evaluating you is where interviews are lost.

Coril exists to close that gap. Not with scripts. Not with cheating tools. With practice that feels real enough that the actual interview feels like your second time, not your first. You do not need months of preparation. You need a few sessions that change how you show up.

What makes Coril different

It reads the actual job posting

Paste a job URL or description. Coril extracts the company, role, seniority level, industry, and requirements. Every question comes from what that specific employer will actually ask. Not a generic question bank. Not random prompts. Questions built from the posting you are preparing for.

It adapts to the role

A senior finance role at an enterprise company gets different treatment than a junior teaching position at a startup. The AI calibrates question difficulty, follow-up depth, and scoring thresholds based on structured signals extracted from the job posting.

Seniority, industry, competency type, and pressure signals all shape how the interview runs.

Five personas, five rounds

Recruiter Screen, Hiring Manager, Behavioral, Skills Assessment, and Final Round. Each persona is built from research into how that round actually runs: what questions come first, when they push back, what they are scoring you on.

The recruiter checks fit. The hiring manager checks depth. The behavioral round checks structure. The skills round finds the edge of your knowledge. The final round reads the person.

Practice in your interview language

Coril detects the language of the job posting you paste and runs the interview in it. English (US and UK), Spanish, French, German, and Dutch. Native voices, native filler-word recognition, native scoring rubric. A French nurse practicing for a Paris hospital gets a French recruiter. A Spanish account manager gets a Spanish hiring manager. The interview matches the language of the real conversation you are preparing for, not an English persona reading translated copy.

Coaching adapts to the round

Real-time coaching after every answer. The recruiter round flags rambling past 90 seconds. The behavioral round checks if your STAR answer is action-heavy. The skills round notes when you are guessing instead of admitting a gap. Different rounds test different things. The coaching matches.

Scoring against the job, not a generic rubric

The evaluation scores your answers against the actual requirements from the posting. If the job asks for FP&A experience and you give an FP&A example, that scores higher than a generic one. If the role is senior and your answers show entry-level depth, the report says so.

Named frameworks for the questions nobody else names

STAR works for past-tense stories. But most interview questions are not behavioral, and most competitors only teach STAR. We name the frameworks that actually fit each question type.

STAR for past-tense behavioral.
PAR for federal and government.
AAAE for situational and conditional ("how would you handle").
STEER for leadership tradeoffs.
ICAO for legal ethics.
MAP for marketing attribution.
ALSO for customer service de-escalation.
CALM for messy-exit framing.
DTS for take-home defense.
SPINE as the delivery scaffold that wraps any of them.
KSA-PAR for federal panels.
FRAME for body language.
FRESH for first-ever interviews.
Bridge-STAR for career changers.

Fourteen named frameworks. The AI thinks in them. It speaks plain English to you. The blog teaches each one with worked examples and the research behind it.

Metrics that match what real interviews measure

Every voice session ends with an animated score ring, a verdict badge, and a performance profile across the things real interviewers notice: filler words per minute, speaking pace calibrated to your role tier, answer-length adherence per round, specificity (named projects, numbers, dates), framework coverage on behavioral answers, I-ownership versus we-deflection, quantified-outcome rate, and generic-adjective count.

A practice plan picks the two or three highest-leverage fixes from your session and names them in plain English. After three sessions, your personal trend chart shows filler words trending down, specificity trending up, framework coverage improving across your last five sessions. Progress against yourself, never against strangers.

Anti-patterns the AI catches as you speak: apologetic openers, hedging pile-ups, generic self-description ("hardworking, dedicated, team player"), corporate buzzwords without evidence, "we" without naming what you specifically did, toxic-job labels, memorized-perfect delivery. You hear the nudge after the answer and a stronger version using your own words.

Per-competency rollup so you know the category, not just the question

Every answer gets tagged to one of the round's five named competencies. Behavioral has STAR coverage, action specificity, quantified outcomes, individual "I" not "we", and conciseness. Final round has self-awareness, candidate questions, cultural fit, why-hire-you closing, and trajectory. The rollup tells you "your STAR coverage is 2.5 of 5" rather than "you scored 3/5 on question two." Decision-useful, not just descriptive.

Time-tagged moments you can mentally re-listen to

Coril surfaces up to six moments per session where filler density spiked, you stalled mid-sentence, or your pace dropped or rushed, each with an M:SS timestamp and a snippet of what you were saying. "Filler spike at 0:47, four fillers in eight seconds" beats a generic note that you used too many fillers.

Question-type patterns across your history

After three sessions, the dashboard surfaces which question types you score lowest on across history: past-tense behavioral, conditional, leadership tradeoff, why-leaving, candidate questions, and ten more. Drill the right type next, instead of guessing at random.

Take the report with you

Download the full session report as a PDF: verdict, score, every answer with the stronger version, performance metrics, recommended next step. Or grab the markdown transcript for your own notes or to paste into ChatGPT for follow-up coaching. The artifact is yours to keep.

Smart retry that varies the questions

Hit Try again and the AI knows you practiced this round before. It varies its opening, probes the topics you under-developed last time, and asks new angles. Retries feel productive instead of repetitive.

The deliberate-practice loop between sessions

Smart Retry varies the questions on a redo. The deliberate-practice loop goes one step further. Each new session auto-loads the previous evaluation's specific growth area into the persona's internal prompt as a practice target, and the interviewer naturally steers toward that gap by the second or third candidate turn — without naming the prior session aloud, without breaking realism. The post-session report shows the prior gap, the score delta (e.g., 5.3 to 6.1, +0.8), and a verdict word (Improving, Stayed flat, or Regressed). Practice between sessions becomes targeted, not random — and the targeting is visible to you, not hidden mechanical optimization.

Free users see one stronger-version answer rewrite per evaluation, picked from the lowest-scored answer in the session. Interview Pass unlocks the full set across every answer. The diagnosis is free; the prescription path is paid.

Quick or Full sessions, not one-size-fits-all

Two session lengths. Quick runs 3 substantive questions in about 4 minutes for a fast baseline before bed, between meetings, or the morning of an interview. Full runs 5 to 7 questions in 8 to 10 minutes for a reliable score and a comprehensive evaluation when you have the time.

First-time users default to Quick so the activation moment lands fast: signup to scored report in roughly five minutes, not fifteen. Returning users keep their last-used length. The score for a Quick session is honestly framed as a baseline, not a verdict, because three answers are noisier than seven.

Practice, not cheating

Some AI interview tools feed you hidden answers during live calls. Companies are catching on. Candidates who rely on those tools get exposed the moment a follow-up question goes off-script.

Coril is the opposite. We make you good enough that you do not need help during the real thing. The skill stays with you after the interview, after the job, after the career change.

It transfers to pitches, meetings, negotiations, and every conversation where you need to think on your feet. For a detailed breakdown of how Coril compares to using ChatGPT for interview practice, we wrote an honest comparison.

Built for every industry

Every interview prep tool is built for software engineers. Coril is built for everyone. Nurses, accountants, teachers, financial analysts, project managers, lawyers, retail managers. If the job has a description, you can practice for it.

Our blog covers interview preparation across healthcare, finance, education, government, legal, retail and hospitality, real estate, sales, operations, customer service, marketing, skilled trades, leadership, and every role in between. 70+ research-backed guides, free for everyone. The blog is the education. The product is the practice. The AI inside the practice tool reads from the same library you read from on the blog.

Who it is for

You have an interview coming up and you want to walk in ready. You are searching across multiple companies and need to practice different rounds for different roles.

You are changing careers and need to frame your experience for a new industry. You are a nurse, a teacher, an accountant, and every interview tool you have found is built for tech.

You are not looking for a tool that gives you the answers. You are looking for practice that makes the answers yours.

Pricing

Free to start. 3 voice sessions per month across all 5 interview rounds.

$15 to pass your next interview. 14 days of unlimited voice sessions, per-answer scoring with answer rewrites, question-type patterns across sessions, time-tagged delivery moments, application tracker, PDF + markdown report download, and Adaptive CV — the interviewer plugs into your uploaded resume and references your actual work history mid-call instead of asking from the job posting alone. Pay once per interview. Re-up when the next one lands. No subscriptions. A single career coaching session costs $75-250.

Climate

Coril contributes 1% of every purchase to Stripe Climate's portfolio of verified carbon removal projects. The percentage is small and the company is small, so the absolute contribution stays modest for now.

The principle is that the percentage never reaches zero. Not when revenue is tight, not before the product is profitable, not later when it is. A profitable software product can commit some back to the systems it depends on. Coril does, from the first dollar.