Intelligent Content
Generation Infrastructure

Context-aware · Self-verifying · Standards-aligned

built by Bohmi.com

An API-first engine for generating educational content at scale

lesson.services is production-grade infrastructure for intelligent content generation. It orchestrates multiple large language models through a resilient pipeline — routing requests across providers with automatic failover, circuit-breaker protection, and structured output validation.

Every generation run flows through a multi-stage pipeline: content construction, self-verification against educational standards, and artifact assembly. The system tracks every token, every cost, every decision — giving operators complete visibility into what was generated, how, and at what price.

Powering content across four production platforms

AskBohmi.com

Conversational AI tutoring with dynamically generated, context-sensitive lesson materials

CTEPro.com

Career and technical education content aligned to industry standards and certification pathways

InKluna.com

Inclusive, differentiated content for special education — adapted for every learner profile

ClassFabric.com

General education lesson planning with standards-aligned, classroom-ready materials

Built for precision, not approximation

Context-Based

Client-supplied context, not generic prompts

Every generation is grounded in rich context provided by the calling application — subject matter, grade level, standards, learning objectives. The engine never guesses what to teach.

Self-Checking

Multi-pass verification pipeline

Generated content passes through structured validation before delivery. Output shape, educational accuracy, and standards alignment are verified programmatically — not left to chance.

Standards-Aligned

Bloom's taxonomy and beyond

Content is structured against Bloom's cognitive levels, mapped to state and national standards, and leveled for appropriate reading complexity. Every artifact is pedagogically intentional.

Multi-Model

Intelligent routing with circuit breakers

Requests are dispatched across Claude and GPT with automatic failover. A Redis-backed circuit breaker with half-open recovery ensures the system self-heals without operator intervention.

Observable

Per-site token accounting and cost tracking

Every LLM call is logged with provider, model, token counts, and calculated cost. Usage is aggregated per site, per day — giving operators and clients full transparency.

Available for content development companies

lesson.services exposes a RESTful API for programmatic content generation. Authenticate with a site-scoped API key, submit generation requests with your content context, and poll for completed artifacts — articles, assessments, translations, and exports.

# Create a generation run
POST /api/v1/generations

# Headers
X-LS-API-Key: ls_live_...
Content-Type: application/json

# Submit blocks, check job status, retrieve artifacts
POST /api/v1/generations/{id}/blocks
GET  /api/v1/jobs/{job_id}
GET  /api/v1/artifacts/{artifact_id}
API key authentication scoped per site
Async job pipeline with status polling
Structured JSON responses throughout
Translation and PDF export endpoints