For two years, every L&D buyer demo we ran ended with the same line: "This is great, but we already have an LMS — what makes you different?" The honest answer was always uncomfortable. Most of what's sold as "AI for L&D" today is a quiz engine bolted onto a content library, with a chatbot at the bottom of every screen. The actual hard parts — defining competencies the org cares about, surfacing skill gaps that have business impact, assessing oral fluency in a regulated profession, certifying compliance training that holds up to an audit — those parts were untouched.

So we stopped describing iTutor as "an AI tutor with a corporate plan" and rebuilt the L&D surface from scratch. Today we're shipping the result: a complete Learning & Development platform that runs the full corporate L&D lifecycle — from competency definition to live oral examination to white-label certification — gated behind three new corporate SKUs. This post walks through what's inside, how the pieces fit together, and what kind of L&D team each tier is for.

The thesis: L&D is not e-learning with a different label

Most LMS-style products treat corporate training as a slightly more compliance-heavy version of K-12 e-learning: pick a course, take a quiz, get a tick, move on. That works for one-off compliance training (anti-bribery, data privacy, fire safety) where the stakes are "did the employee click through the slides," but it fails for everything else.

Real L&D in a real company looks different. A commercial-banking team needs to know that a relationship manager can actually walk a corporate client through a credit covenant in Arabic, in real time, without reading from a script. A pharmaceutical sales team needs to know that a new rep can answer a physician's objection about a competitor's drug without misstating efficacy data. A pre-hospital ambulance service needs to know that a paramedic can verbally talk through a STEMI protocol while their hands are on a patient. None of those are "did you read the slides" questions — they're behaviour, language, and judgement under pressure.

That's the gap the iTutor L&D platform was built for. The product wraps four hard things into one workflow: competency frameworks the business actually cares about, AI-delivered learning sessions in twelve languages including Egyptian-dialect Arabic, voice-based oral competency exams with anti-AI proctoring, and defensible certification with frozen-per-cert white-label branding. Underneath all of it is the same machinery that powers iTutor's K-12 and higher-ed products — the curriculum builder, the voice agent, the analytics pipeline — repurposed for the corporate side.

Competency frameworks: the part most L&D platforms skip

A competency framework is the spine of any serious L&D programme. It defines the behaviours, knowledge, and skills the organization expects from each role — broken down into proficiency levels, mapped to departments, scored against rubrics. Without one, "training" is just content distribution. With one, every learning hour ties back to a measurable shift in capability.

Most L&D buyers we talk to either don't have a framework, have one that lives in a Word document nobody opens, or have one in an HRIS that the LMS can't read. iTutor handles all three cases:

Manual builder — a structured editor for L&D admins to define departments → categories → requirements → levels by hand. Tree-style with soft delete and archived state, so old structure doesn't break historical reports.
PDF framework import — upload an existing job-architecture document, role-profile bundle, or certification standard and the AI extracts the competency tree directly. The pipeline reads the file, infers the hierarchy, and seeds your tenant's framework with editable nodes. Available on Growth and Enterprise tiers.
Per-org versioning and metering — every framework, level, and requirement is scoped to your tenant, with quotas on how many you can create per tier (1 framework on Starter, 5 on Growth, unlimited on Enterprise).

Once the framework exists, every other surface in the product hangs off it: learning sessions are scoped to specific levels, gate exams test specific requirements, certificates are issued per level, and analytics roll up by department.

Sessions, scenarios, and twelve languages of voice

An iTutor learning session is not a video. It's a live, two-way voice conversation with an AI tutor that has read your competency framework and your uploaded corporate materials — SOPs, product specs, compliance docs, sales playbooks — and conducts a structured tutoring session against a specific level. The learner can interrupt, ask follow-ups, and request examples. The AI adapts difficulty based on prior answers.

Three things make the session engine specifically L&D-grade rather than a generic chatbot:

Twelve voice languages with regional dialect support. English, Arabic (including Egyptian dialect — not just MSA), French, Spanish, German, Portuguese, Italian, Dutch, Turkish, Indonesian, Malay, and Urdu. Voice languages are tier-gated: Starter is English-only, Growth adds Arabic, Enterprise unlocks all twelve. For MENA-region deployments, dialect-correct Arabic is the difference between a tool people use and a tool people abandon after week one.
Scenarios. Beyond straight-tutoring sessions, learners can run branching scenarios — promotion-readiness simulations (Specific Level scenarios), and on Growth/Enterprise, career-shift scenarios that let an employee in role A simulate a session as if they were already in role B, scored against B's competency profile. This is internal-mobility programme infrastructure, not training videos.
Real-time barge-in and prosody. The voice agent handles natural interruption, doesn't talk over the learner, and matches conversational pacing rather than reciting at a fixed cadence. Latency is low enough that a confident learner can hold an actual back-and-forth.

The L&D Competency Exam: oral assessment with anti-AI proctoring

This is the part of the product we spent the longest on, and the one we think no other corporate L&D platform offers honestly today.

The L&D Competency Exam is a voice-based oral examination conducted by an AI examiner. The candidate joins a webcam-and-mic session, the examiner introduces the exam in the candidate's chosen language, and proceeds to ask context-aware questions drawn from the level's question pool — adapting follow-ups based on the candidate's prior answers, exactly like a human examiner would. Every answer is transcribed, every video frame is paired to its acoustic signal, and the entire session is replayable for HR and L&D reviewers afterwards.

The examiner is not a chatbot in a window. It runs on the same low-latency voice stack as iTutor's tutoring product, with natural barge-in, mic muting during AI thinking time (so background noise doesn't bleed into the transcript), and per-language prosody. A candidate taking the exam in Arabic gets an examiner that speaks Egyptian-dialect Arabic with appropriate examination register — formal but not robotic.

Why "anti-AI" proctoring is the table stakes nobody talks about

Here's the inconvenient truth about voice-based corporate exams in 2026: a candidate with a smartphone, a second monitor, and a free LLM can passably fake competence in real time. They speak the question into the LLM, read the answer back to the examiner, and pass. If your certification programme is going to mean anything to a regulator, an auditor, or a hiring manager, you need to make that attack expensive — ideally impossible.

iTutor's Growth and Enterprise tiers ship a five-layer anti-AI probe stack:

Acoustic probes. A real-time detector listens for cough, clap, and hum signatures — used as periodic challenge prompts the candidate must respond to. An LLM-relayed candidate fails these because the human in front of the screen isn't actually listening to the examiner's audio in real time.
Multi-display detection. Browser-side checks flag when a second display is connected to the candidate's machine, when the focus leaves the exam window for too long, or when developer tools are opened. Each event is logged with a timestamp into the proctoring report.
Voiceprint speaker similarity. The candidate's voice is fingerprinted at session start and continuously compared against the talking voice throughout the exam. A switch to a different speaker (a colleague taking over off-camera) is detected and flagged.
Paired frame and signal events. Every video frame is stored with a paired acoustic signal so reviewers can verify lip-sync — important when reviewing whether a candidate was actually speaking versus replaying recorded answers from a phone.
Real-time probe responses. Periodic micro-challenges the candidate must respond to within a tight window — designed to be trivial for a present human and impossible for an LLM-relay loop.

Probe events appear in the affect-analysis report alongside requirement-level scores, so HR and L&D reviewers see exactly what triggered. Starter tier ships basic proctoring (camera + transcript audit) only — appropriate for low-stakes pilot deployments. Growth and Enterprise are where the full probe stack lives.

Affect analysis: the report HR actually reads

Once an exam ends, iTutor produces a deep examiner-style affect-analysis report — the document that reviewers, hiring managers, and compliance auditors actually consume. Highlights:

Per-requirement narrative analysis. Not a numeric score with a thumbs-up. A paragraph of examiner-grade prose for each requirement explaining what the candidate said, where they were strong, where they were weak, and what specifically they would need to study to close the gap.
Top-of-page weak-topics summary. Reviewers see the candidate's three weakest topics first, with topic-level recommendations.
No-face card. Periods where the candidate's face was off-camera are surfaced as a dedicated card, not buried in a log.
Probe-details card. Every triggered probe (cough, clap, multi-display, voiceprint mismatch) is listed with its timestamp and the surrounding context.
Single-source aggregate score. One canonical pass/fail or band score, computed deterministically across chunks so the same input always produces the same output. No drift between runs.

The narrator pipeline is chunked across multiple LLM calls so a long exam doesn't get summarised down to "candidate did okay" — every requirement gets its own examiner-quality paragraph, with the actual student transcript pulled into answer snippets rather than rubric rationale leaking back as the candidate's words.

White-label certificates that hold up to an audit

Certificates are the part of L&D that gets ignored until the day a regulator asks for one. iTutor's certificate engine is built around a single non-negotiable: a certificate must always render the way it was issued, even if the org rebrands afterwards. We do this by snapshotting the org's branding into the certificate at issuance time.

L&D admins on the Enterprise tier manage their certificate branding in /corporate/ld/settings/branding. The editor lets them upload a logo, set primary, secondary, and accent colours, and write English and Arabic issuer and footer text. The next certificate that issues writes a copy of the branding into a branding_snapshot_json column on the certificate row itself. Three years from now, when an auditor asks to see Sarah's 2026 compliance certificate, it renders with Sarah's company's 2026 colours and 2026 logo — even if the company has since rebranded.

The same snapshot drives the public verification page at /public/ld/verify/{token}. A regulator or hiring manager scanning a printed certificate's QR code lands on a verify page that renders in the issuing org's branding, displays the candidate name, the level certified, the framework, the issue date and expiry, and the cryptographic verification of the certificate token. No iTutor branding leaks onto the regulator's screen.

Other certificate features:

Per-level certification. An employee progressing through Levels 1 → 2 → 3 of a framework gets a certificate for each level, not just one at the end.
Expiry tracking for recurring training. Compliance certs (safety, regulatory, code-of-conduct) get explicit expiry dates with manager-notify hooks before they lapse.
Approval gates. Certificates can be configured to require a human reviewer to approve the exam before the certificate is issued. The reviewer queue is its own role on Growth and Enterprise.
Conditional approvals. A reviewer can approve "with conditions" — the candidate gets a provisional certificate and a follow-up training assignment that must be completed within a window before the cert finalises.

Mentorship, content proposals, and the manager-notify loop

Three smaller-but-important capabilities round out the workflow side of the product:

Mentorship assignments. Structured mentor↔mentee pairings with employee-manager history tracking. A mentee's progress and exam outcomes flow into their mentor's dashboard automatically.
Content proposals. Learners and SMEs can submit proposed competency content (a new question for a question pool, a new scenario branch, a corrected requirement) into a reviewer queue. Approved proposals merge into the framework.
Manager notification on milestones. A manager gets a notification when a direct report passes a level, fails a gate exam, or has a certificate about to expire. Configurable per-org.

Mentorship and content proposals are Growth and Enterprise tier features. Manager-notify is on all three tiers.

The three SKUs and who they're for

The L&D platform is sold as three corporate SKUs. Existing academic corporate plans (the corporate_starter / growth / enterprise tiers we already had for schools and universities) do not unlock the L&D surface — to reach competency frameworks, gate exams, certificates, and scenarios, an org must hold one of the new ld_* SKUs.

L&D Starter — $299/month ($2,990/year)

The pilot tier. Designed for an L&D team that wants to evaluate the platform on a single team — a department of 25 employees, one framework, English-only voice agent, basic proctoring. 200 live exam minutes per month, 100 gate attempts, 25 certificates. No anti-AI probes, no PDF framework import, no white-label certs, no SSO. Community support. The point is to let an L&D buyer prove value internally on a real team before committing to Growth.

L&D Growth — $999/month ($9,990/year)

The primary tier. This is what most mid-market HR and L&D teams will sign. 250 employees across one or two departments, 5 frameworks at 12 levels each, 2,000 live exam minutes per month, 1,500 gate attempts, 500 certificates per month, 20 concurrent live exams, English plus Arabic voice. Full anti-AI proctoring stack (cough/clap/hum probes, multi-display detection, voiceprint similarity). PDF framework import. Mentorship, content proposals, reviewer role. Promotion, specific-level, and career-shift scenarios. Cohort benchmarks analytics. Bulk enrolment. 90-day audit log retention. Priority email support.

L&D Enterprise — $2,999/month ($29,990/year)

The top tier, for whole-company L&D programmes. 2,500 employees, 25 admins, 250 teachers, 50 reviewers. Unlimited frameworks. 20,000 live exam minutes per month, 15,000 gate attempts, 5,000 certificates, 100 concurrent live exams. All twelve voice agent languages. White-label certificate templates with frozen-per-cert branding. SSO via SAML/OIDC. iTest pre-exam binding for orgs that already run iTest deployments. Competency timeline analytics. 365-day audit log retention. Dedicated success manager. The Enterprise tier is sales-led — no Stripe or Paymob self-checkout; the front-end renders a contact form that creates a custom-request row our team picks up within one business day.

Self-checkout for Starter and Growth is Paymob-only — no Stripe on the L&D corporate tiers. Per-org metering is enforced server-side: live exam minutes, gate attempts, PDF imports, certificates issued, and concurrent live exams are all metered, with quotas resetting at the start of each calendar month. Exceeding a quota returns a structured upgrade hint that the front-end renders as an in-app upgrade chip pointing to the next tier.

Integrations, residency, and the boring-but-important compliance surface

L&D buyers ask the same five questions every time. Here's where iTutor lands on each:

Existing LMS. SCORM 1.2 and 2004, LTI 1.3, and xAPI for round-tripping with Cornerstone, SuccessFactors, Docebo, Workday Learning, and any other compliant LMS.
SSO. SAML 2.0 and OIDC, both on the Enterprise tier.
Data residency. Per-region deployment in EU, US, and MENA. Important for regulated industries and for MENA-headquartered enterprises with sovereign-cloud requirements.
iTest integration. For orgs already running iTest, pre-exam binding pulls cohorts and pushes results back into the corporate iTest system. Enterprise tier only.
Audit trail. Every action against the L&D surface is logged with user, timestamp, and resource. Retention is 30 days on Starter, 90 on Growth, 365 on Enterprise. Logs are queryable from the admin console and exportable for external audit.

What this looks like for a real L&D team — three concrete cases

Case 1: a mid-sized commercial bank, 800 employees, MENA

The relationship-banking team needs every RM to demonstrate that they can walk a corporate client through a credit covenant in Arabic, in real time, without a script. The team starts on Growth ($999/month), defines a "Commercial Banking RM" framework with five levels (Junior → Senior), uploads the bank's product specs and credit-policy documents, and runs every RM through a Level 3 oral examination in Arabic at the end of their first quarter. Anti-AI probes catch the two RMs who tried to relay the questions to ChatGPT on a second monitor. The certificates issue with the bank's logo and Arabic footer text frozen into each one.

Case 2: a regional pharmaceutical sales force, 350 reps, EU

A new product launch requires every rep to demonstrate they can answer five common physician objections without misstating efficacy data. The team uses Growth, builds a "New Product Launch — Q3" framework with one level, uploads the FDA approval summary and the physician objection-handling playbook, and runs a 15-minute oral exam per rep. The affect-analysis report flags reps who hedged on efficacy or misquoted contraindications, with timestamped transcript snippets. Reps who fail get a follow-up scenario session targeted at the specific objection they fumbled, then re-examine within seven days.

Case 3: a multinational manufacturer, 4,500 employees, global

The L&D team is rolling out a unified leadership development programme across operations in twelve countries. They start on Enterprise ($2,999/month) for the language coverage and the white-label requirement (the company has strong brand standards and will not issue any certificate that doesn't render in their corporate colours). Eight frameworks roll out in parallel — one per business unit. Anti-AI probes are mandatory. SSO via SAML to the corporate IdP. iTest integration pushes all results back into the existing performance-management system. The dedicated success manager runs the rollout cadence with the head of L&D for the first six months.

How to start

The L&D plans are live on itutor.study/pricing — switch to the L&D tab to compare Starter, Growth, and Enterprise side by side. Starter and Growth are self-checkout via Paymob; Enterprise is sales-led through the contact form.

If you're an existing iTutor customer on a corporate academic plan and you want to add the L&D surface, contact your account manager — we'll provision an L&D tier alongside your existing org without disrupting your current users. If you're new to iTutor, the simplest path is the L&D Starter tier as a pilot — set up a single team, define one framework, run a real cycle, then decide whether to expand to Growth.

For a deeper look at the underlying assessment engine, the How AI Tutoring Actually Works post covers the voice stack and the latency engineering that makes natural-feeling oral conversation possible. For the certificate-verification side, the public verify endpoint at /public/ld/verify/{token} is browseable — you can scan any test certificate and see exactly what an external auditor or hiring manager would.

L&D has been the slowest-moving slice of corporate-software buying for a decade. Most teams are still buying tools that treat training like e-learning — a content library and a quiz. We think the next ten years will reward platforms that treat L&D as what it actually is: behaviour change, language fluency, and judgement under pressure, certified in a way that holds up to an audit. The iTutor L&D platform is our bet on that future. We'd be glad to walk any L&D team through a live demo — book one from /contact or sign up for the Starter tier today.

We Built a Complete L&D Platform — Not Another Quiz Engine. Here's What's Inside.