ARCHITECT.
THE WORLD'S FIRST GENERATIVE ENGINE FOR CURRICULUM ARCHITECTURE. TURN RAW DATA INTO PRODUCTION-READY KNOWLEDGE INFRASTRUCTURE IN SECONDS. ZERO HALLUCINATION. ENTERPRISE INTEGRATION. TOTAL CONTROL.
Full structured timeline from one API call
Quality gates before any curriculum ships
Median generation time under full queue load
Cost on cache-hit — identical requests are free
Not a wrapper.
Infrastructure.
Every layer production-hardened. Built to run under load, not collapse under it. Hover any card for the full technical spec.
AI Orchestrator
Every generation passes through a sequential 4-layer pipeline: Planner → Logic Filter → Structural Critic → AI Critic. Nothing ships unless all four agree.
Planner defines scope. Logic Filter enforces prerequisite ordering. Structural Critic validates module-lesson ratios. AI Critic runs semantic hallucination detection.
Depth-Mapped Output
Module count scales automatically with your requested timeframe. Every lesson has a title, objective, estimated duration, and difficulty calibration.
A 6-month curriculum generates 9–12 modules. A 4-week sprint generates 3–4. Depth is never sacrificed for speed — generation time scales, not quality.
BullMQ Async Queue
Your POST returns a job ID in under 200ms. Generation runs entirely in the background via BullMQ workers, concurrency-controlled per API key.
No request ever blocks your thread. Workers are isolated per tenant. Queue depth, retry logic, and dead-letter handling are all production-configured.
Semantic Cache
Semantically equivalent requests — even if phrased differently — hit the Redis cache. Zero AI tokens consumed. Zero cost on repeat requests.
Cache keys are derived from a normalized semantic vector of your request. 7-day TTL. Cache invalidation available via the API. Cost savings are logged per key.
API Key Gating
Every route is protected by issued API keys. Rate limits, quota thresholds, and role assignments are enforced at the middleware layer before any AI call.
Keys are scoped per organization. Quota is tracked in real-time. Keys can be revoked, rotated, or rate-limited without touching your integration.
Edit & Explain API
One endpoint rewrites any lesson or entire module using a natural-language instruction. Another explains any concept at any depth, for any audience.
PATCH /curriculum/:id/lesson/:lessonId with {instruction: 'Make this more Socratic'}. GET /explain with {topic, audience, depth} returns structured pedagogical breakdown.
Five steps.
Full curriculum.
Send a POST with your topic, timeframe, and complexity. The engine queues, generates, validates through 4 AI gates, and fires your webhook — all in under 10 seconds.
POST /v2/curriculum/generate HTTP/1.1Host: api.mentrast.comx-api-key: mk_live_7f91a3bc9204ee1dContent-Type: application/json{ "user_id": "usr_7f91a3bc", "topic": "Advanced Node.js Microservices", "intent": "Production-grade backend architecture mastery", "timeframe": "6 months", "complexity": "advanced", "options": { "diagnostic_first": true, "include_estimates": true, "output_format": "structured_json", "webhook_url": "https://your-app.com/hooks/mentrast" }}Who uses the engine.
If you build, sell, or deliver knowledge products — this is your infrastructure layer.
Embed production-grade curriculum generation in your product.
Stop hand-authoring course outlines. One API call returns a structured, AI-validated curriculum tailored to your learner's complexity level and timeframe. Plug it into your existing UX.
Deploy skill-mapped curriculum pathways across your entire workforce.
Map training topics to job roles and timeframes. The engine generates adaptive, role-specific curriculum structures that your L&D team can deploy at any scale — without manual authoring overhead.
Automate course creation. Focus on delivery.
Instructional designers spend 80% of their time on structure. Let the engine handle that in seconds, and redirect your team's expertise toward facilitation, coaching, and live sessions.
Generate full bootcamp curricula on demand.
Advanced complexity mode generates diagnostic-first, expert-level curricula. Every module starts with an architectural gap analysis. No fluff. No beginner filler. Pure, deep progression.
No public pricing.
All custom. All yours.
The API is not free. Every plan requires a key. Pricing is scoped to your volume, usecase, and team size — reach out and we'll build a plan together.
Starter
For small teams and early-stage products integrating curriculum generation into their platform.
Growth
For scaling EdTech platforms and enterprise L&D teams that need volume, reliability, and advanced features.
Enterprise
For large institutions, white-label products, and SaaS companies who need SLAs, custom integrations, and dedicated infrastructure.
Pay As You Go
Not ready for a monthly commitment? Get billed per curriculum generated. Same API, same quality, same key. Usage is tracked and invoiced monthly.
Your key.
Our engine.
THE GENERATIVE ENGINE FOR CURRICULUM ARCHITECTURE IS AVAILABLE TO QUALIFIED ENTERPRISE TEAMS. REQUEST ACCESS BELOW.
MENTRAST CURRICULUM ENGINE · © 2026