1·01Foundation
45mLinear Algebra for AI
Vectors, matrices, eigen-decomposition and why they power neural nets.
OMS AI University
The AI-native learning platform
engineered for a 2050 workforce.
An expandable knowledge platform organizing every core AI discipline — foundations, research, engineering, automation, business, security and future technology — into a single scalable curriculum.
7
Learning tracks
210+
Topic blocks
∞
Continuously expanding
210 topics shown
AI-100 · Track
AI Foundation
Math, ML and neural network fundamentals for every OMS practitioner.
1·02Foundation
50mProbability & Statistics
Distributions, Bayes rule, sampling and estimation for ML models.
1·03Foundation
40mCalculus for Optimization
Derivatives, gradients and the chain rule behind backpropagation.
1·04Foundation
35mInformation Theory
Entropy, KL-divergence and cross-entropy losses explained.
1·05Foundation
30mIntro to Machine Learning
Supervised, unsupervised and reinforcement learning at a glance.
1·06Foundation
45mLinear & Logistic Regression
Baseline models every AI engineer must master before deep learning.
1·07Foundation
50mDecision Trees & Ensembles
Random forests, gradient boosting and XGBoost intuition.
1·08Foundation
55mNeural Networks 101
Perceptrons, activations and universal approximation.
1·09Applied
60mBackpropagation Deep Dive
Derive gradients step-by-step for a small feed-forward net.
1·10Applied
60mConvolutional Networks
How CNNs learn spatial features for vision tasks.
1·11Applied
55mRecurrent Networks & LSTMs
Sequence modelling before the transformer era.
1·12Applied
70mTransformers From Scratch
Attention, multi-head, positional encoding — the modern backbone.
1·13Foundation
35mTokenization & Embeddings
BPE, subword vocab and vector semantics.
1·14Foundation
30mRegularization & Overfitting
Dropout, weight decay, augmentation and early stopping.
1·15Applied
40mOptimizers: SGD to AdamW
Momentum, adaptive rates and modern training recipes.
1·16Advanced
45mLoss Landscapes
Why deep nets generalize — flat minima and implicit bias.
1·17Foundation
30mDatasets & Benchmarks
MNIST, ImageNet, GLUE, MMLU — what they actually measure.
1·18Foundation
25mEvaluation Metrics
Precision, recall, F1, ROC-AUC and their business meaning.
1·19Foundation
25mBias-Variance Tradeoff
Diagnosing under- vs over-fitting like a scientist.
1·20Applied
35mFeature Engineering
The unglamorous work that still beats bigger models.
1·21Applied
30mCross-Validation Strategies
K-fold, stratified, time-series and leakage traps.
1·22Applied
40mGPU Foundations
Tensors, CUDA, mixed precision and memory budgets.
1·23Applied
60mPyTorch Essentials
Modules, autograd, dataloaders and training loops.
1·24Advanced
45mJAX for the Curious
Functional autodiff and XLA for research-grade speed.
1·25Applied
40mHugging Face Ecosystem
Transformers, datasets, accelerate and the hub.
1·26Foundation
25mPrompting 101
Zero-shot, few-shot and chain-of-thought foundations.
1·27Foundation
30mLanguage Model History
From n-grams to GPT-4 — a 30-minute tour.
1·28Applied
35mVector Search Basics
Embeddings, ANN indexes and similarity metrics.
1·29Foundation
30mEthics of AI Foundations
Fairness, transparency, accountability from day one.
1·30Foundation
20mCareer Map for AI Builders
Roles, skills and portfolios in the OMS AI economy.
AI-200 · Track
AI Research
Frontier models, alignment and reproducibility for OMS scientists.
2·01Research
30mReading AI Papers Efficiently
A repeatable framework to digest arXiv in minutes.
2·02Research
45mScaling Laws
Compute, data and parameter scaling for language models.
2·03Research
40mEmergent Abilities
What appears — and doesn't — as models scale.
2·04Advanced
60mRLHF & DPO
Aligning language models with human preferences.
2·05Research
40mConstitutional AI
Anthropic's approach to self-supervised alignment.
2·06Research
55mMechanistic Interpretability
Circuits, features and reading a model's mind.
2·07Research
50mSparse Autoencoders
Decomposing activations into human-readable features.
2·08Advanced
45mMixture of Experts
Routing, load balancing and trillion-parameter economics.
2·09Research
45mState Space Models
Mamba, S4 and the post-transformer frontier.
2·10Advanced
60mDiffusion Models Deep Dive
Score matching, DDPM, DDIM and modern samplers.
2·11Applied
45mLatent Diffusion & Stable Diffusion
How to compress image generation into consumer GPUs.
2·12Research
45mVideo Diffusion
Temporal consistency, Sora-class architectures.
2·13Advanced
50mMultimodal Foundation Models
CLIP, Flamingo, GPT-4V design patterns.
2·14Research
45mWorld Models
Learned simulators for planning and robotics.
2·15Advanced
55mReinforcement Learning Recap
Policy gradients, PPO, Q-learning, actor-critic.
2·16Research
50mModel-Based RL
Dreamer, MuZero and planning-inside-latents.
2·17Advanced
45mReasoning & Chain-of-Thought
Self-consistency, tree-of-thought, verifier models.
2·18Advanced
40mProgram-Aided LMs
Tool use, code interpreters and symbolic reasoning.
2·19Advanced
45mLong Context Techniques
Rope scaling, attention sinks, retrieval-augmentation.
2·20Research
35mSpeculative Decoding
Draft-and-verify inference for faster generation.
2·21Advanced
40mQuantization Research
GPTQ, AWQ, FP8 and 1-bit LLMs.
2·22Advanced
40mDistillation & Compression
Turning frontier models into edge-ready students.
2·23Research
40mContinual Learning
Avoiding catastrophic forgetting in production models.
2·24Research
45mMeta-Learning
MAML, in-context learning and learning-to-learn.
2·25Research
45mNeurosymbolic AI
Combining logic and neural nets for verifiable reasoning.
2·26Research
40mAI for Science
AlphaFold, materials, climate — the new lab bench.
2·27Research
45mRobotics Foundation Models
RT-2 and generalist robot policies.
2·28Applied
30mReproducibility Playbook
Seeds, envs, containers, model cards and datasheets.
2·29Research
35mWriting an AI Paper
Structure, ablations, honest limitations.
2·30Foundation
25mOMS Research Charter
How OMS Organization contributes to open AI science.
AI-300 · Track
AI Engineering
Production LLM systems, agents and the AI Developer OS toolchain.
3·01Applied
40mLLM App Architecture
Prompts, retrieval, tools, memory and guardrails.
3·02Applied
35mPrompt Engineering Patterns
Role, context, examples, constraints and reflection.
3·03Applied
30mStructured Outputs
JSON mode, tool calling and schema-validated generation.
3·04Applied
35mFunction & Tool Calling
Turning an LLM into a controller of real APIs.
3·05Applied
55mRetrieval-Augmented Generation
End-to-end RAG with chunking, reranking, evaluation.
3·06Applied
30mChunking Strategies
Fixed, semantic, hierarchical and late-chunking.
3·07Applied
30mReranking with Cross-Encoders
Boosting relevance with a small dedicated model.
3·08Applied
35mVector Databases
pgvector, Qdrant, Weaviate, LanceDB compared.
3·09Applied
35mHybrid Search
BM25 + vectors + metadata filters done right.
3·10Applied
45mLLM Evaluation
Golden sets, LLM-as-judge, task-specific scorers.
3·11Applied
35mObservability for LLMs
Traces, spans, replay and cost telemetry.
3·12Applied
30mStreaming UX
Token streams, tool-call UX, cancellation.
3·13Advanced
55mAgent Design Patterns
ReAct, Planner-Executor, Reflexion, multi-agent.
3·14Advanced
50mMulti-Agent Systems
Blackboards, roles, arbitration and cost control.
3·15Advanced
45mLong-Term Agent Memory
Episodic, semantic and procedural memory stores.
3·16Applied
40mGuardrails & Safety Layers
Input/output filters, PII, jailbreak defenses.
3·17Applied
40mCost & Latency Optimization
Caching, batching, routing and model tiers.
3·18Applied
30mModel Routing
Choosing the cheapest model that still meets the bar.
3·19Applied
35mFine-Tuning vs RAG vs Prompting
A decision framework for real production teams.
3·20Applied
50mLoRA & QLoRA Fine-Tuning
Cheap adapters for private domain knowledge.
3·21Advanced
45mServing Open Models
vLLM, TGI, TensorRT-LLM in production.
3·22Applied
40mEdge & On-Device AI
llama.cpp, MLX, WebGPU and local-first agents.
3·23Applied
40mAI Developer OS
The OMS toolchain for shipping agentic products.
3·24Applied
35mVector Embeddings at Scale
Batch pipelines, incremental updates, drift.
3·25Advanced
40mMulti-Tenant AI Apps
Isolation, quotas and per-tenant knowledge bases.
3·26Advanced
45mRealtime Voice Agents
Speech-to-speech pipelines with low latency.
3·27Applied
40mVision Pipelines
OCR, layout parsing, visual QA in production.
3·28Applied
30mData Contracts for AI
Trustworthy inputs for trustworthy outputs.
3·29Applied
35mAI Testing Strategy
Unit, snapshot, canary and regression suites.
3·30Applied
30mShipping an AI Feature
From idea to safe rollout in 30 days.
AI-400 · Track
AI Automation
Agentic workflows and the Mother AI operating model at enterprise scale.
4·01Foundation
25mWorkflow Automation 101
Triggers, actions, branches — the automation atoms.
4·02Advanced
45mAgent Orchestration
Coordinating dozens of agents without chaos.
4·03Applied
40mMother AI Operating Model
How OMS orchestrates specialist agent divisions.
4·04Applied
35mBPMN for Agentic Ops
Modeling AI-augmented business processes.
4·05Advanced
40mEvent-Driven Agents
Queues, streams and reactive automation.
4·06Applied
30mHuman-in-the-Loop Design
When to ask, when to act, when to escalate.
4·07Applied
30mAI for Sales Ops
Lead scoring, outreach and CRM enrichment agents.
4·08Applied
35mAI for Marketing Ops
Content, SEO, ads and analytics agents.
4·09Applied
35mAI for Customer Support
Deflection, triage and quality assurance.
4·10Applied
40mAI for Finance Ops
Invoicing, reconciliation and anomaly detection.
4·11Applied
30mAI for HR & People Ops
Sourcing, screening and onboarding automation.
4·12Applied
35mAI for Legal Ops
Contract review, clause extraction, redlining.
4·13Applied
30mAI for Procurement
Vendor scoring, PO automation and spend intelligence.
4·14Applied
40mAI for Supply Chain
Demand forecasting and routing optimization.
4·15Applied
35mAI for Media Production
Editing, captioning and distribution pipelines.
4·16Applied
40mAI for E-commerce
Merchandising, pricing and personalization.
4·17Applied
35mAI for Analytics
Text-to-SQL, dashboards and narrative reports.
4·18Applied
30mRobotic Process Automation + AI
Combining RPA with LLMs for unstructured work.
4·19Advanced
40mBrowser Automation Agents
Playwright, computer-use and the web-as-API.
4·20Advanced
40mVoice Automation
IVR replacement with realtime voice agents.
4·21Applied
35mDoc Intelligence Pipelines
PDFs, forms and tables at enterprise scale.
4·22Applied
30mKnowledge Base Automation
Auto-curating truth from meetings and tickets.
4·23Advanced
30mAgent Marketplaces
Publishing, pricing and governing shared agents.
4·24Foundation
20mAutonomy Ladder
L0-L5 automation and where humans still belong.
4·25Applied
30mSLOs for Agents
Latency, accuracy and cost service-level objectives.
4·26Advanced
35mCost-Aware Orchestration
Budgets, back-pressure and graceful degradation.
4·27Applied
30mChange Management for AI
Rolling out agents without breaking teams.
4·28Applied
25mMeasuring Automation ROI
Cycle time, quality, deflection and margin.
4·29Advanced
35mFailure Modes of Agents
Loops, drift, tool errors and how to recover.
4·30Applied
35mOMS Automation Blueprint
Reference architecture for AI-native operations.
AI-500 · Track
AI Business
Product strategy, monetization and the AI-native operating playbook.
5·01Applied
35mAI Product Strategy
Choosing wedge, moat and defensibility in the AI era.
5·02Applied
30mAI Business Models
Seat, usage, outcome and workforce pricing patterns.
5·03Applied
30mPricing AI Products
Metering, margins and cost pass-through.
5·04Applied
30mGTM for AI Startups
PLG, sales-led and hybrid motions.
5·05Advanced
40mEnterprise AI Sales
Security, compliance and champion enablement.
5·06Advanced
30mAI Category Design
Naming the game you intend to win.
5·07Applied
25mAI Positioning
Framing narrative in a noisy market.
5·08Applied
30mAI Content Marketing
Compounding attention with expert-driven media.
5·09Applied
25mCommunity-Led Growth
Building an ecosystem around your platform.
5·10Applied
30mAI Partnerships
Data, distribution and model provider alliances.
5·11Applied
35mFundraising for AI
What investors actually diligence in 2026.
5·12Advanced
35mUnit Economics of AI
Gross margin math when the COGS is inference.
5·13Applied
30mOrg Design for AI Companies
Small teams, big leverage, AI colleagues.
5·14Applied
25mHiring an AI Team
Roles, ladders and evaluation rubrics.
5·15Applied
25mVendor Strategy
Model, cloud and tooling procurement.
5·16Applied
30mData as Competitive Advantage
Proprietary data flywheels and feedback loops.
5·17Applied
35mLegal & IP for AI
Model licensing, training data and output rights.
5·18Applied
30mPublic Policy Landscape
EU AI Act, US EO, UK, Bangladesh, ASEAN.
5·19Applied
25mAI Insurance & Risk
New coverage models for AI-driven operations.
5·20Advanced
30mM&A in the AI Era
Buy-vs-build for models, data and talent.
5·21Advanced
30mAI Board Governance
What boards should ask about AI strategy.
5·22Foundation
25mExecutive AI Literacy
A one-hour brief for non-technical leaders.
5·23Applied
30mAI-Native Consulting
Selling outcomes, not hours.
5·24Applied
40mVertical AI Playbooks
Media, finance, health, logistics, retail.
5·25Applied
30mOMS Ventures Model
How OMS launches and scales AI-native ventures.
5·26Advanced
35mEcosystem Strategy
Turning products into platforms into economies.
5·27Applied
30mGlobal Expansion
Localization, data residency and partnerships.
5·28Applied
25mSustainability & AI
Energy, water and responsible scaling.
5·29Applied
25mAI Brand Building
Trust as the ultimate distribution channel.
5·30Advanced
35mThe 2030 Founder Playbook
How to build a defensible AI-native company.
AI-600 · Track
AI Security
Responsible AI, red-teaming, zero-trust deployment and governance.
6·01Applied
40mThreat Modeling for AI
STRIDE, LINDDUN and AI-specific attack surfaces.
6·02Applied
40mPrompt Injection
Direct, indirect and cross-tool injection attacks.
6·03Applied
35mJailbreak Defenses
Layered controls, canaries and detectors.
6·04Applied
35mData Exfiltration Risks
How agents leak — and how to stop them.
6·05Applied
30mPII Handling
Detection, redaction and residency for AI systems.
6·06Applied
30mSecrets Management
Keys, tokens and per-tenant isolation.
6·07Advanced
40mZero-Trust for Agents
Least-privilege tool access and just-in-time creds.
6·08Advanced
35mSupply Chain Security
Model, dataset and package provenance.
6·09Research
30mModel Watermarking
Tracking outputs and provenance signals.
6·10Advanced
45mAdversarial ML
Evasion, poisoning, extraction and membership attacks.
6·11Advanced
40mDifferential Privacy
Formal privacy for training and analytics.
6·12Advanced
40mFederated Learning
Training across silos without moving raw data.
6·13Advanced
35mConfidential Compute
Enclaves and encrypted inference basics.
6·14Applied
40mGovernance Frameworks
NIST AI RMF, ISO 42001, EU AI Act mapping.
6·15Applied
25mModel Cards & Datasheets
Transparency artifacts that scale trust.
6·16Applied
35mBias & Fairness Testing
Group, individual and counterfactual measures.
6·17Advanced
40mRed-Teaming LLMs
Structured adversarial evaluation programs.
6·18Applied
25mContent Provenance (C2PA)
Signed media in an AI-generated world.
6·19Applied
30mDeepfake Detection
Signals, limits and organizational response.
6·20Applied
30mAI Incident Response
Playbooks for model failures and misuse.
6·21Applied
35mAI Auditing
Internal and third-party assurance patterns.
6·22Advanced
40mRegulated Deployments
Health, finance and public sector guardrails.
6·23Applied
30mAccess Control for AI Apps
RBAC, ABAC and per-record policies.
6·24Applied
30mLogging & Forensics
Immutable audit trails for agent actions.
6·25Foundation
25mResponsible AI Principles
OMS AI Charter for teams and partners.
6·26Foundation
30mAI Ethics Case Studies
Learning from real-world failures.
6·27Applied
30mHuman Oversight Design
Meaningful control without decision-paralysis.
6·28Advanced
30mSafe Autonomy Levels
Guardrails per capability, per risk tier.
6·29Applied
25mAI Insurance & Liability
Emerging contracts and coverage.
6·30Applied
30mOMS Trust Architecture
How OMS operationalizes responsible AI.
AI-700 · Track
AI Future Technology
AGI, robotics, quantum-AI and the 2050 research agenda.
7·01Research
45mPaths to AGI
Scaling, algorithmic and hybrid perspectives.
7·02Research
45mSuperalignment
Aligning systems smarter than their overseers.
7·03Research
35mAgentic Economies
When agents transact with other agents.
7·04Advanced
40mDigital Twins
Simulating factories, cities and organizations.
7·05Research
45mHumanoid Robotics
General-purpose bodies for general-purpose brains.
7·06Advanced
40mEmbodied AI
Perception-action loops in the physical world.
7·07Advanced
40mAutonomous Vehicles 2.0
End-to-end learned driving stacks.
7·08Research
35mBrain-Computer Interfaces
Non-invasive and invasive BCI landscape.
7·09Research
35mNeuromorphic Computing
Spiking networks and event-driven silicon.
7·10Research
40mQuantum Machine Learning
Where quantum actually helps AI (and doesn't).
7·11Research
30mPhotonic AI Chips
Optical compute for post-transistor scaling.
7·12Advanced
30mMemory-Centric AI
In-memory compute and HBM4 futures.
7·13Research
40mAI + Biotech
Protein design, cell engineering and drug discovery.
7·14Research
35mAI + Climate
Modeling, mitigation and adaptation.
7·15Research
30mAI + Energy
Grid optimization and fusion research assistants.
7·16Research
35mAI + Materials
Autonomous labs and self-driving discovery.
7·17Research
30mAI + Space
On-orbit autonomy and deep-space missions.
7·18Research
35mSynthetic Biology + AI
Designing life with foundation models.
7·19Research
30mLongevity Research
AI-accelerated aging science.
7·20Research
30mAI for Governance
Simulation, policy analysis and civic tech.
7·21Research
35mPost-Labor Economics
What society optimizes for when work is optional.
7·22Research
25mUniversal Basic Compute
Compute as a right, not a privilege.
7·23Applied
30mPersonal AI
Life-long, private, on-device companions.
7·24Research
35mAI Cities
Sensing, planning and governing at metro scale.
7·25Applied
30mAI in Education 2050
One tutor per learner, forever.
7·26Applied
30mAI in Healthcare 2050
Continuous, preventive, personalized care.
7·27Applied
30mAI in Media 2050
Generative studios and audience-of-one experiences.
7·28Applied
30mAI in Finance 2050
Autonomous treasuries and machine-native markets.
7·29Research
40mExistential Risk & Safety
Sober analysis of long-horizon AI risk.
7·30Foundation
30mOMS 2050 Roadmap
How OMS Organization contributes to the long future.
A learning platform designed to keep growing.
OMS AI University is engineered as a scalable block system. New topics, courses, research briefs and certifications are added continuously across all seven tracks — a permanent home for the AI-native workforce.