Software Bill of Materials (SBOM) for the ibl.ai Platform
SBOM, software bill of materials, generative AI platform, LLM-agnostic, LangChain, Langfuse, Flowise, OpenAI GPT-4, Google Gemini, Azure OpenAI, Anthropic Claude, AWS Bedrock, open-source LMS, OpenAPI, Python SDK, JavaScript SDK, OAuth2, OIDC, SAML, LTI 1.3, ReactJS, Next.js, React Native, MentorAI, university CIO, edtech, AI tutor, permissive licenses, vendor lock-in avoidance, cost control, enterprise security, higher education technology
The ibl.ai platform is a generative-AI-powered learning system built on an open-source LMS foundation, extended with a flexible LLM layer and fully exposed through OpenAPI-compliant services. All core components use permissive licenses (MIT / Apache 2.0), ensuring zero hidden licensing costs and no vendor lock-in for the institution. The architecture is modular, standards-based, and designed for secure campus deployment on-prem or in any cloud.
1 · Generative AI Engine & Frameworks
| Component | Role in Platform | License | |-----------|------------------|---------| | LangChain | Framework for building and chaining LLM-powered tools; powers tutoring agents, content generation, and multi-model orchestration. | MIT | | Langfuse | Observability & tracing layer for LLM calls; enables prompt/response logging, performance dashboards, and debugging. | MIT | | Flowise | No-code visual builder for LLM workflows and agents; accelerates rapid prototyping and custom AI flows. | Apache 2.0 | | OpenAI SDK (Python & Node) | Official libraries for GPT's latest models; supports streaming, fine-tuning, and advanced usage analytics. | MIT | | Google Gemini SDK | Unified client for Gemini models on Vertex AI; offers multimodal (text + image) generation and enterprise controls. | MIT |2 · Supported Large Language Models
| Provider | Example Models | Highlights | |----------|----------------|------------| | OpenAI | Latest Available Models | Leading accuracy, broad ecosystem, coding & conversation excellence. | | Google Cloud AI | Gemini | Native multimodal reasoning, Vertex AI integration, fine-tune workflows. | | Microsoft Azure | Azure OpenAI | Enterprise compliance, regional data residency, Azure AD integration. | | Anthropic | Latest Claude Models | Safety-focused “Constitutional AI,” 100k-token context for long documents. | | AWS Bedrock | Amazon Titan (+ third-party models) | Flexible mix-and-match models under AWS security and cost controls. | > Model-agnostic: Administrators may choose, combine, or swap models without code changes.3 · Platform Core (LMS & API)
| Component | Description | License | |-----------|-------------|---------| | Open-Source LMS Core | Full course delivery, enrollment, grading, and analytics engine. Mature, scalable, and extensible to meet university requirements. | Permissive OSS | | REST API (OpenAPI) | 100 % feature coverage via OpenAPI-defined endpoints; supports content, tutoring, analytics, and admin operations. | — | | Python SDK | Auto-generated client; simplifies server-side integrations and data pipelines. | MIT | | JavaScript / TypeScript SDK | Auto-generated client for web/mobile apps and serverless functions. | MIT | | Auth Layer | OAuth2 / OIDC & SAML2 for SSO, plus LTI 1.3 for cross-LMS embedding; supports server-to-server or client-initiated flows. | — |4 · Front-End & Application Ecosystem
| Framework / App | Purpose | License | |-----------------|---------|---------| | ReactJS | Core library for dynamic web UIs (dashboards, portals). | MIT | | Next.js | Full-stack React framework with server-side rendering and API routes. | MIT | | React Native | Cross-platform mobile framework (iOS & Android). | MIT | | MentorAI (reference app) | Pre-built AI tutor (web + mobile) showcasing best practices; code shareable with the university. | Source available | | Custom Partner Apps | Any partner-built web/mobile apps leveraging the OpenAPI & SDKs; authenticate via OAuth2/SSO. | OSS frameworks |5 · Licensing & Cost Perspective
* All components use permissive licenses (MIT / Apache 2.0 / AGPLv3). * No proprietary runtime fees; the university retains full code ownership. * Costs arise only from optional usage-based AI model calls with chosen providers. * Modular design allows on-prem, cloud, or hybrid deployment while meeting security and compliance requirements.6 · At-a-Glance Benefits for CIOs
| Pillar | Value Delivered | |--------|-----------------| | Open & Extensible | OpenAPI endpoints, open-source code, flexible SDKs. | | Vendor-Neutral AI | Swap or mix LLM providers without lock-in. | | Enterprise Security | OAuth2/OIDC, SAML, LTI 1.3, and role-based access. | | Future-Proof | Rapid adoption of new models via LangChain & Flowise. | | Cost Control | No platform license fees; pay only for chosen AI usage and infra. |Related Articles
Microsoft Copilot + ibl.ai: Building an AI stack universities actually own
Microsoft Copilot excels as a GPT-4 assistant baked into Microsoft 365, yet it lacks the course-grounding, data residency, and model flexibility campuses require. ibl.ai’s open, LLM-agnostic mentorAI backend supplies that secure layer—RAG over syllabus content, multi-tenant SOC 2/FERPA controls, analytics, and big cost savings—so universities keep Copilot’s front-line productivity while owning the AI core.
How mentorAI Integrates with Anthropic
mentorAI lets universities route each task to Anthropic’s Claude 3 family through their own Anthropic API key or AWS Bedrock endpoint, sending high-volume chats to Haiku (≈ 21 k tokens per second), deeper tutoring to Sonnet, and 200 k-context research queries to Opus—no code changes required. The platform logs every token, enforces safety filters, and keeps transcripts inside the institution’s cloud, while Anthropic’s commercial-API policy of not using customer data for training protects FERPA/GDPR compliance.
How ibl.ai Supercharges Khan Academy’s Mission—Without Competing
Khanmigo offers GPT-4-powered, student-friendly tutoring on top of Khan Academy’s content, but campuses still need secure ownership, LMS/SIS integration, and model flexibility. ibl.ai’s mentorAI supplies that backend—open code, LLM-agnostic orchestration, compliance tooling, analytics, and cost control—letting universities embed Khanmigo today, swap models tomorrow, and run everything inside their own cloud without vendor lock-in.
Claude + ibl.ai: A Blueprint for AI-Native Universities
Anthropic’s new Claude for Education supplies the guarded, Socratic chat front end, while ibl.ai’s share-the-code MentorAI delivers the back-office muscle—LLM-agnostic orchestration, SSO/LTI, audit logs, and faculty overrides—inside a university-owned cloud. Together they ground Claude in syllabus files, blend models, monitor costs, and swap engines at will, eliminating lock-in.