AS300 — AI Platform Engineer
Intensive 3-month curriculum to master Red Hat OpenShift AI (RHOAI), LLM/ML model deployment, MLOps/LLMOps pipelines, and GenAI applications in production.
399 hours (57 days)
57 business days
Red Hat Certified Specialist in AI/ML (EX267)
January 11, 2027
→ April 02, 2027
€7,490 incl. VAT
20% VAT included
Curriculum Overview
The AS300 — AI Platform Engineer curriculum is an intensive 399-hour program to master the deployment, scaling, and operation of enterprise AI platforms on Kubernetes and Red Hat OpenShift AI.
Methodology: From container orchestration to distributed LLM serving (vLLM, KServe) and model alignment with InstructLab.
This curriculum covers 8 official Red Hat courses (including AI267 and AI296 specifically focused on AI), a certification preparation phase, a 105-hour final project on a complete GenAI use case, and a professional defense before a jury.
First cohort (Winter 2027): Alan Turing (January 11, 2027 → April 02, 2027).
Learning Objectives
Upon completing the AS300 curriculum, you will be able to:
- Master Red Hat OpenShift AI (RHOAI) for AI/ML model deployment in production
- Deploy and serve generative models (LLMs via vLLM) and predictive models (OpenVINO) on OpenShift
- Build data and model pipelines (MLOps/LLMOps) with Kubeflow Pipelines
- Design and deploy GenAI applications (RAG, agents, guardrails) with Llama Stack
- Monitor AI models with TrustyAI (fairness, reliability, drift)
- Train and fine-tune LLMs with Red Hat Enterprise Linux AI (RHEL AI) and InstructLab
- Obtain the Red Hat EX267 certification (Red Hat Certified Specialist in AI/ML)
Target Audience
Designed for profiles looking to specialize in artificial intelligence infrastructure:
- Career-transition profiles with technical or scientific backgrounds (mathematics, data, physics, engineering) aiming for AI infrastructure roles
- Post-graduate students and recent graduates seeking high-demand operational skills bridging cloud platforms and LLMOps
- IT professionals and software developers looking to operate GPU clusters and deploy enterprise AI models
- Data Engineers and MLOps aiming to deploy, supervise, and secure production models
Prerequisites
To join the AS300 curriculum, you must:
- Working comfort with the Linux command line : navigation, files and permissions, users, processes, editing with vim/nano, and using
sudo - Basic container concepts: comprehensive OpenShift training is included in the curriculum
- Have basic networking knowledge: TCP/IP, DNS, HTTP
Software Versions & Ecosystem
This curriculum is aligned with the Aperta Scientia internal release 2026.08-3, synchronized with the latest Red Hat OpenShift AI versions and cutting-edge open source GenAI ecosystem components:
| Course / Module | Official Title | Target Software Version | Role in the Architecture |
|---|---|---|---|
| DO188 | Podman & Containers | Podman v5.2+ | Container foundation and GPU runtimes |
| DO180 / DO280 | OpenShift Admin I & II | RHOCP v4.22 / v4.18+ | Enterprise Kubernetes infrastructure |
| DO288 | OpenShift Development | RHOCP Pipelines & GitOps | Application deployment automation |
| AI267 | Red Hat OpenShift AI (RHOAI) | RHOAI v3.3 (GA) | vLLM model serving, MLOps pipelines, JupyterLab |
| AI296 | GenAI Fundamentals & InstructLab | InstructLab v0.22+ / RHEL AI 1.3+ | Model alignment and IBM Granite 3.0 |
| KServe / vLLM | Model Serving Engine | vLLM v0.8+ / KServe v0.14+ | LLM inference with PagedAttention & multi-GPU |
| Vector Stack | RAG & Embeddings | Milvus v2.5+ / ChromaDB v0.6+ | Vector indexing and RAG pipelines |
| TrustyAI | AI Governance & Safety | TrustyAI v1.32+ | Guardrails, fairness metrics and explainability |
Detailed Curriculum
Global Structure — 399 hours / 57 days
| # | Course | Module Title | Days | Hours |
|---|---|---|---|---|
| 00 | BOOTCAMP | RHLS Bootcamp + Linux Fundamentals | 2 d | 14 h |
| 01 | DO188 | Podman & Containers | 3 d | 21 h |
| 02 | DO180 | Kubernetes & OpenShift Basics | 5 d | 35 h |
| 03 | DO280 | OpenShift Administration I | 5 d | 35 h |
| 04 | DO380 | OpenShift Administration II | 5 d | 35 h |
| 05 | DO288 | OpenShift Development | 5 d | 35 h |
| 06 | DO322 | OpenShift Installation Lab | 2 d | 14 h |
| 07 | AI267 | Red Hat OpenShift AI (RHOAI) | 5 d | 35 h |
| 08 | AI296 | GenAI Fundamentals + Granite + RHEL AI | 3 d | 21 h |
| 09 | PREP | Intensive Certification Preparation (EX267) | 5 d | 35 h |
| 10 | EXAM | Official Red Hat EX267 Certification Exam | 1 d | 7 h |
| 11 | PROJECT | Hands-on Team AI & MLOps Platform Project | 15 d | 105 h |
| 12 | JURY | Professional Jury Defense & Evaluation | 1 d | 7 h |
| Total curriculum | 57 days | 399 hours | ||
Red Hat Courses (8 courses)
231h (58%)
Bootcamp, cert prep & jury
63h (16%)
Engineering Project
105h (26%)
Course / Project Ratio
74 / 26
Module 0 — RHLS Bootcamp & Linux Fundamentals (14h / 2 days)
Bootcamp & Linux Fundamentals
🎯 Learning Objective
Objective: Navigate the Linux filesystem, edit files, and get started with the Red Hat Learning Subscription (RHLS) environment.
📚 Detailed Chapter Syllabus
- Pt.1 Content: RHLS platform onboarding
- Pt.2 Linux command line (navigation, file manipulation)
- Pt.3 Text editing (vim/nano)
- Pt.4 User management, privileges (`sudo`), and permissions
Module 1 — DO188: Podman & Containers (21h / 3 days)
Podman & Containers
🎯 Learning Objective
Objective: Build, run, and manage isolated containers with Podman, and ensure data persistence.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Introduction and Overview of Containers
- Ch.2 Podman Basics: manage and run containers
- Ch.3 Container Images: navigate registries
- Ch.4 Custom Container Images: build with Containerfile
- Ch.5 Persisting Data: database containers
- Ch.6 Troubleshooting Containers: logs and remote debugger
- Ch.7 Multi-container Applications with Podman Compose
- Ch.8 Container Orchestration with OpenShift and Kubernetes
- Ch.9 Compreview
Module 2 — DO180: Kubernetes & OpenShift Basics (35h / 5 days)
Kubernetes & OpenShift Basics
🎯 Learning Objective
Objective: Deploy, manage, and troubleshoot containerized applications on an OpenShift cluster.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Introduction to Kubernetes and OpenShift
- Ch.2 Kubernetes and OpenShift CLI and APIs
- Ch.3 Run Applications as Containers and Pods
- Ch.4 Deploy Managed and Networked Applications on Kubernetes
- Ch.5 Manage Storage for Application Configuration and Data
- Ch.6 Configure Applications for Reliability
- Ch.7 Manage Application Updates
- Ch.8 Compreview
Module 3 — DO280: OpenShift Administration I (35h / 5 days)
OpenShift Administration I
🎯 Learning Objective
Objective: Configure and administer vital components of an OpenShift cluster.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Declarative Resource Management (YAML/Kustomize)
- Ch.2 Deploying Packaged Applications (Helm)
- Ch.3 Authentication and Authorization (RBAC/HTPasswd)
- Ch.4 Network Security (NetworkPolicies)
- Ch.5 Exposing non-HTTP/SNI Applications
- Ch.6 Enabling Developer Self-service
- Ch.7 Managing Kubernetes Operators (OLM)
- Ch.8 Application Security (SCC)
- Ch.9 OpenShift Updates
- Ch.10 Compreview
Module 4 — DO380: OpenShift Administration II (35h / 5 days)
OpenShift Administration II
🎯 Learning Objective
Objective: Automate "Day 2" operations and configure monitoring.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Authentication and Identity Management (LDAP/OIDC)
- Ch.2 Backup, Restore, and Migration (OADP)
- Ch.3 Pod Scheduling
- Ch.4 OpenShift GitOps (ArgoCD)
- Ch.5 OpenShift Monitoring (Prometheus/Grafana)
- Ch.6 Logging for Red Hat OpenShift
- Ch.7 Cluster Partitioning
- Ch.8 Compreview
Module 5 — DO288: OpenShift Development (35h / 5 days)
OpenShift Development
🎯 Learning Objective
Objective: Design, build, and deploy cloud-native applications optimized for OpenShift.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Red Hat OpenShift for Developers
- Ch.2 Deploying Simple Applications
- Ch.3 Building and Publishing Container Images
- Ch.4 Managing Red Hat OpenShift Builds
- Ch.5 Managing Red Hat OpenShift Deployments
- Ch.6 Deploying Multi-container Applications (Templates, Helm, Kustomize)
- Ch.7 Continuous Deployment with OpenShift Pipelines (Tekton)
- Ch.8 Comprehensive Review
Module 6 — DO322: OpenShift Installation Lab (14h / 2 days)
OpenShift Installation Lab
🎯 Learning Objective
Objective: Plan and execute the automated installation of a complete OpenShift cluster.
📚 Detailed Chapter Syllabus
- Pt.1 Content: IPI and UPI installation, on cloud provider and virtualized environment, post-installation validation
Module 7 — AI267: Red Hat OpenShift AI — RHOAI (35h / 5 days)
Red Hat OpenShift AI — RHOAI
🎯 Learning Objective
Objective: Deploy AI environments, build data pipelines, and serve MLOps/LLMOps models.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Introduction to Red Hat OpenShift AI: RHOAI as a platform for building, deploying, and monitoring AI models
- Ch.2 Using Workbenches for AI/ML Development: create and manage Jupyter environments
- Ch.3 Fundamentals of Model Serving (KServe): deployment modes, model routes, resource parameters
- Ch.4 Serving Generative AI Models (LLMs via vLLM) and Predictive Models (OpenVINO)
- Ch.5 Monitoring AI Models with TrustyAI: fairness metrics, drift detection, reliability
- Ch.6 Introduction to Data Science Pipelines (Kubeflow): automate AI/ML workflows
- Ch.7 Advanced Kubeflow Pipelines Development: containerized components, artifacts, experiment tracking
- Ch.8 GenAI Model Selection, Optimization, and Evaluation: RHOAI model catalog, Red Hat AI repository
- Ch.9 Building GenAI Applications: RAG, agents, Llama Stack
Module 8 — AI296: GenAI Fundamentals + Granite + RHEL AI (21h / 3 days)
GenAI Fundamentals + Granite + RHEL AI
🎯 Learning Objective
Objective: Train, optimize, and evaluate language models (LLMs) using Red Hat tools.
📚 Detailed Chapter Syllabus
- Pt.1 Content by chapter:
- Ch.1 Generative AI Fundamentals: Capabilities, Challenges, Models, and Techniques
- Ch.2 Granite Models for Enterprise Generative AI: positioning, use case suitability evaluation
- Ch.3 Training Large Language Models with Red Hat Enterprise Linux AI (RHEL AI): InstructLab, fine-tuning, synthetic data generation
- Ch.4 Deploying Models with Red Hat Enterprise Linux AI: vLLM as inference runtime, model release
Module 9 — Certification Preparation (35h / 5 days)
Red Hat EX267 Certification Preparation
🎯 Learning Objective
Objective: Consolidate technical skills and prepare for official exam conditions.
📚 Detailed Chapter Syllabus
- Pt.1 Content: Intensive review targeting the EX267 (Red Hat Certified Specialist in AI/ML) exam objectives, building on the Administration (EX280) and Development (EX288) foundations and the AI-specific components
- Pt.2 Troubleshooting workshops
- Pt.3 Timed mock exams with real constraints (no external internet)
Module 10 — Certification Exams (7h / 1 day)
Official Red Hat EX267 Certification Exam
🎯 Learning Objective
Objective: Obtain Red Hat professional certifications.
Module 11 — Final Project (105h / 15 days)
Hands-on AI & MLOps Engineering Project & Agile Rituals
Production-grade team implementation: AI architecture scoping, OpenShift AI deployment, MLOps/LLMOps pipelines, and final jury defense.
Hands-on Team AI & MLOps Platform Project
🎯 Learning Objective
Objective: Design, deploy, and secure a complete AI use case (MLOps/LLMOps) from data to final application.
📚 Detailed Chapter Syllabus
- Pt.1 Timeline:
- → Setup (D1–D3): Framing, team formation, GitOps init, OpenShift provisioning, RHOAI deployment & tooling
- → Sprint 1 (D4–D6): Data ingestion/preparation, training/fine-tuning (Workbenches), MLOps pipelines
- → Sprint 2 (D7–D9): Model Serving (vLLM, OpenVINO), application integration (RAG, Agents), inference pipeline
- → Sprint 3 (D10–D11): Monitoring (TrustyAI), Guardrails, security hardening, performance
- → Finalization (D12–D14): Deliverables, documentation, Code Freeze, defense dry-runs
- → D15: Final preparation
Module 12 — Jury & Validation (7h / 1 day per group)
Professional Jury Defense & Evaluation
🎯 Learning Objective
Objective: Demonstrate the acquisition of curriculum competencies during a professional defense.
📚 Detailed Chapter Syllabus
- Pt.1 Process: Formal defenses before a jury of professionals
- Pt.2 Q&A
- Pt.3 Post-mortem
Results Indicators & Performance Commitments
Newly Registered Organization: In accordance with professional training regulations, Aperta Scientia is a newly declared training organization (NDA
11 78 89002 78). Consolidated statistical performance data will be published upon completion of the first cohort.
| Performance Indicator | 2027 Target | Methodology & Source |
|---|---|---|
| Attendance & Completion Rate | ≥ 90 % | Digital attendance verification via Edusign (399h) |
| Overall Learner Satisfaction | ≥ 85 % | End-of-course satisfaction surveys |
| Employer / Sponsor Satisfaction | ≥ 85 % | Manager & sponsor evaluations |
| Red Hat Certification Pass Rate | ≥ 75 % | Official Red Hat certification exams |
| Job Placement at 6 Months | ≥ 80 % | Post-graduation 6-month placement survey |
| Drop-out Rate | ≤ 5 % | Continuous monitoring & support |
Target Technologies & Versions
Official technologies taught in version 2026.08-3 of the curriculum.
Red Hat OpenShift AI (RHOAI)
v3.3 / v2.25+Full enterprise AI/ML platform & Llama Stack Operator (AI267)
vLLM
v0.8+ / v0.7+High-performance LLM serving engine & PagedAttention
InstructLab
v0.22+LAB method & open source LLM fine-tuning (AI296)
RHEL AI & Granite Models
Granite 3.0/3.1 & RHEL AI 1.3+Enterprise foundation models and specialized inference server environment
KServe & ModelMesh
v0.14+Serverless multi-model serving & GPU autoscaling
OpenVINO Model Server
v2025.1+Hardware acceleration and CPU/GPU/NPU optimizations
Kubeflow Pipelines
v2.4+Reproducible data and model workflow orchestration
LangChain & LlamaIndex
LangChain v0.3+ / LlamaIndex v0.12+AI agent orchestration frameworks & advanced RAG pipelines
Vector Databases
Milvus 2.5+ / Chroma 0.6+ / pgvectorHigh-performance semantic vector storage & similarity search
TrustyAI
v1.32+Explainability, bias monitoring & ethical AI guardrails
PyTorch
v2.6+ / v2.5+Industry standard Deep Learning framework for training and inference
2027 Sessions & Cohorts
Intensive 399-hour programs delivered over 59 working days including 3 days of individual jury defenses. Recommended registration deadline at J-28 for funding and administrative processing. Schedule is provisional and subject to change*.
Promotion Ada Lovelace
AS300-2027-WIN-RMT-FR-01
Key Milestones
Cohort Alan Turing
AS300-2027-WIN-RMT-EN-01
Key Milestones
Promotion Claude Shannon
AS300-2027-SPR-RMT-FR-02
Key Milestones
Cohort John von Neumann
AS300-2027-SPR-RMT-EN-02
Key Milestones
Promotion Margaret Hamilton
AS300-2027-AUT-RMT-FR-03
Key Milestones
Cohort Larry Wall
AS300-2027-AUT-RMT-EN-03
Key Milestones
* Scheduled dates are provisional and subject to change based on cohort schedules.
Practical Information & Details
Training Objectives
Detailed in the pedagogical curriculum above.
Prerequisites
- ✓ Working comfort with the Linux command line: navigation, files and permissions, users, processes, editing with vim/nano, and using sudo
- ✓ Basic container concepts (comprehensive OpenShift training is included in the curriculum)
- ✓ Basic networking knowledge: TCP/IP, DNS, HTTP
Target Audience
- → Candidates aiming to become AI Platform Engineers
- → Engineers and scientists from other fields targeting a tech career change
- → IT professionals with an OpenShift foundation seeking AI/ML skills
- → Data Engineers or ML Engineers wanting to operate AI platforms on Kubernetes
- → DevOps/Platform Engineers looking to integrate AI into their pipelines
Duration & Schedule
Duration: 399 hours (57 days)
Structure: 57 consecutive working days (excluding public holidays)
Format: Remote — 7 effective hours per day (synchronous virtual classroom and hands-on RHLS labs)
Daily schedule: 9:00 AM – 12:30 PM & 2:00 PM – 5:30 PM (7h / day, 35h / week)
Start date: January 11, 2027
Access Modalities & Timeline
Remote training (interactive virtual classroom and hands-on labs). Registration via contact form on the website or by email at [email protected]. Pre-enrollment positioning interview to validate prerequisites.
Access timeline: within 4 weeks after file and prerequisites validation. First cohort opens January 11, 2027.
2027 Tuition & Funding
2027 Tuition: €7,490 incl. VAT
20% VAT included
Eligible funding: OPCO, skills development plans, corporate training, individual or company funding. Contact us
Teaching & Evaluation Methods
Teaching methods
Active and iterative pedagogy: theory → immediate practice on RHLS cloud labs. End-to-end Data Science use cases on Jupyter notebooks. Prompt engineering workshops, fine-tuning on small datasets, MLOps/LLMOps pipeline deployment. Final team project with agile rituals.
Evaluation modalities
Continuous assessment: Guided Exercises (GE), Compreview quizzes, successful execution of inference pipelines. Final assessment: official Red Hat EX267 (Red Hat Certified Specialist in AI/ML) 100% practical exam, then project defense before a jury of professionals.
Accessibility & Disabilities
Disability Officer ([email protected])
Aperta Scientia is committed to making its training programs accessible to people with disabilities (PwD). Contact us before enrollment so we can work together on any necessary accommodations (venue accessibility, adapted materials, extra time, specific support).
Accessibility contact: [email protected]
Performance & Results Indicators
Satisfaction rate
Data available after the 1st cohort
Success rate
Data available after the 1st cohort
Placement rate
Data available after the 1st cohort
Results indicators will be published after the first cohort (January–March 2027).