Cloudboosta Academy · AI-Native Cloud Programme · Cohort 1

Not an AI course. An infrastructure programme.

Six live Saturdays for working cloud and DevOps engineers. You will build, deploy, govern and automate AI workloads in production, on AWS and Azure, with the tools you will actually use at work. The goal is not a new role. The goal is to be indispensable in the one you have.

Starts
Saturday 7 November 2026
Format
Live online on Google Meet, 10:00 to 13:00 GMT
Length
6 Saturdays, lab and Monday deliverable each week
Cohort
Maximum 20 participants
Reserve a place See the six weeks

£799 in full, or two instalments of £430. A place is confirmed when the first payment is received.

Who this programme is for

Already in a cloud role. Being asked about AI.

This is for people already employed in cloud and DevOps who want to extend into AI infrastructure before it becomes a required skill. It is not suitable for complete beginners.

You will fit if you are
  • A cloud engineer with a year or more of hands-on AWS, Azure or GCP
  • A DevOps engineer who manages pipelines and infrastructure and wants to add AI deployment
  • A platform engineer or SRE being asked by your organisation to support AI adoption
  • A cloud engineer looking to accelerate into a senior or lead role
Prerequisites, at least one of
  • Hands-on AWS, Azure or GCP in a professional context
  • Experience deploying or managing cloud infrastructure beyond tutorial level
  • A working understanding of VPCs, IAM roles, containers and CI/CD pipelines
  • At least one cloud certification at Associate level or above

New to cloud? Start with the Academy pathways and come back for this one.

Curriculum

Six Saturdays. Something to take to work every Monday.

Every week has a live session, a hands-on lab in your own AWS sandbox and n8n instance, and a Monday deliverable: something you take back to your current role immediately.

Session 1Saturday 7 November 2026Week 1

AI for the cloud engineer: the infrastructure perspective

How AI workloads behave differently from traditional cloud applications, and what that means for the engineers who run the infrastructure.

  • How AI workloads differ from traditional applications: compute profiles, latency, cost structures
  • GPU vs CPU: selecting the right instance type for AI workloads on AWS and Azure
  • The AI cloud stack: managed services, build vs buy, and where the abstraction layer sits
  • Key vocabulary for cloud engineers: tokens, inference, embeddings, RAG, fine-tuning
Lab

Deploy an LLM inference endpoint on AWS Bedrock and benchmark latency and cost.

Monday deliverable

An infrastructure assessment of how your current cloud environment would need to change to support an AI workload.

Session 2Saturday 14 November 2026Week 2

n8n: AI workflow automation for cloud engineers

Build AI-powered workflows your team will actually use. n8n is an open-source workflow automation platform that connects AI models to real engineering infrastructure, deployed on your own cloud instance.

  • Introduction to n8n and self-hosted deployment on AWS EC2 or an Azure VM
  • Connecting n8n to OpenAI, AWS Bedrock and other AI APIs
  • Real engineering use cases: AI-powered incident summarisation, log analysis, intelligent alert routing
  • Building a multi-step AI workflow: monitoring alert, AI analysis, Slack notification
Lab

Build and deploy an n8n workflow that connects a cloud monitoring alert to an AI summary.

Monday deliverable

A working n8n workflow solving a real problem in your current environment.

Session 3Saturday 21 November 2026Week 3

Deploying LLMs on cloud infrastructure

From API call to production-grade AI application: the engineering decisions behind running an LLM reliably and cost-effectively at scale.

  • AWS Bedrock, Azure OpenAI Service and Google Vertex AI: a practitioner comparison
  • Production LLM deployment: API gateway, load balancing, auto-scaling and rate limiting
  • Prompt engineering for engineers: reliable, deterministic outputs at scale
  • Cost management: token monitoring, caching strategies, avoiding runaway inference spend
Lab

Deploy a production-style LLM endpoint with a cost and latency monitoring dashboard.

Monday deliverable

A cost model for running an LLM-powered internal tool in your organisation.

Session 4Saturday 28 November 2026Week 4

RAG systems: making AI know what it needs to know

Retrieval-Augmented Generation is the most commercially important AI pattern for enterprise cloud engineers. This week covers the full stack, from data ingestion to production query.

  • What RAG is and why it matters more than fine-tuning for most enterprise use cases
  • The RAG stack: embeddings, vector databases, retrieval layers and cloud data infrastructure
  • Vector database options: pgvector on RDS, Pinecone, Weaviate, managed vs self-hosted
  • Building a data ingestion pipeline for internal knowledge bases
  • n8n integration: automating RAG updates when new documents are added
Lab

Build a RAG system that lets an AI answer questions about a set of internal technical documents.

Monday deliverable

A RAG architecture proposal for one specific internal knowledge problem.

Session 5Saturday 5 December 2026Week 5

AI agents and intelligent automation

From single model calls to autonomous engineering workflows: agents that take actions, make decisions and complete multi-step tasks within defined boundaries.

  • What AI agents are and how they differ from simple LLM calls
  • Agent orchestration patterns in cloud environments: when to use agents and when not to
  • Building agents with n8n: autonomous workflows that make decisions and call APIs
  • Real use cases: auto-remediation, cost optimisation suggestions, automated compliance checking
  • Security and guardrails: preventing runaway automation, audit trails, approval gates
Lab

Build an AI agent that monitors a cloud metric and autonomously suggests or executes a remediation action.

Monday deliverable

An agent design document for one autonomous workflow in your current role.

Session 6 · CapstoneSaturday 12 December 2026Week 6

Production readiness, security, governance, and capstone

Everything that separates a demo from something you can put your name on. Followed by capstone presentations from every participant.

  • Security for AI workloads: IAM permissions, data residency, prompt injection risks, output sanitisation
  • AI governance: audit trails, model versioning, bias monitoring, compliance with AI regulation
  • Cost governance: FinOps principles applied to AI workloads, tagging, budgets, anomaly detection
  • Monitoring AI in production: model drift, inference quality, cost per query
Capstone

Each participant presents the AI integration they built during the programme, with peer and facilitator review and structured feedback.

Then

Certificate and next steps.

Tools and platforms

Production-grade, in active enterprise use.

Nothing here is a teaching toy. Every tool is one you can defend choosing in a design review on Monday.

  • n8nAI workflow automation, self-hosted on an instance provided per participant
  • AWS BedrockManaged LLM deployment on AWS
  • Azure OpenAI ServiceGPT-4 family with enterprise compliance controls
  • pgvector and PineconeVector databases for RAG systems
  • Prometheus and GrafanaExtended for AI workload monitoring
  • AWS CloudWatchCost monitoring and anomaly detection for AI workloads
  • TerraformInfrastructure as Code for every AI stack component
Gbemisola Ojo speaking on stage at Cloud DevOps Connect, London, 2024
Your facilitator

Taught by someone who signs off cloud designs during the week, not someone who reads about them.

Gbemisola Ojo
Cloud DevOps Manager, PwC UK · 4x AWS Certified · Founder, Cloudboosta
  • Manages multiple cloud projects at PwC
  • Teaches the AWS modules on the Academy pathways: high availability, databases, the core services
  • Facilitates all six live sessions and the capstone review
After six weeks

What you will be able to do.

  1. 01

    Deploy and configure AI services on AWS Bedrock and Azure OpenAI at production standard

  2. 02

    Build n8n workflows that automate real engineering tasks using AI

  3. 03

    Design and deploy a RAG system for internal knowledge bases

  4. 04

    Build and govern AI agents within defined safety boundaries

  5. 05

    Estimate and control the cost of running AI workloads in the cloud

  6. 06

    Apply security controls specific to AI workloads

  7. 07

    Monitor AI systems in production: latency, cost and drift

  8. 08

    Present a complete AI integration, built during the programme, to a business or engineering audience

What is included

Everything you need, during and after.

During the programme
  • Six live Saturday sessions, three hours each
  • Session recordings within 24 hours, with 90-day access
  • An AWS sandbox environment for all lab exercises
  • An n8n cloud instance, provisioned, configured and ready for Week 2
  • Weekly lab guides and reference architecture documents
  • Capstone project brief, submission and live review
  • Access to the Cloudboosta Slack community
  • Cloudboosta AI-Native Cloud Engineer Certificate on completion, with a LinkedIn endorsement from Cloudboosta
After the programme
  • 90 days of continued Slack community access
  • 90-day access to all session recordings
  • Priority consideration for the Cloudboosta mentor programme
  • The alumni network: peer connection with every cohort graduate
  • Early access and alumni pricing on future Cloudboosta programmes
Pricing

One price. Two ways to pay.

Places are limited to 20 participants per cohort. A place is confirmed only when the first payment is received.

Full payment
£799
One payment, immediate full access.
Two instalments
£430 × 2
£430 on enrolment and £430 before Week 3, £860 in total.
Corporate
10% off per head
Two or more team members. Corporate invoicing available.
Reserve a place

Tell us where you are. We will send the payment details.

We read every reservation by hand, because the programme is only right for engineers already in a cloud role. Someone from Cloudboosta replies from support@cloudboosta.co.uk, usually within one working day, with payment details or a quick question. Your place is held once the first payment clears.

  • Starts Saturday 7 November 2026, ends Saturday 12 December 2026
  • Live on Google Meet, 10:00 to 13:00 GMT, recorded within 24 hours
  • Maximum 20 participants
Ask a question on WhatsApp

How you would like to pay

Your details go to the Cloudboosta team only and are used to answer this request. See the privacy notice.

Got questions?

Is this a beginner course?
No. It assumes you already work with VPCs, IAM, CI/CD and containers, and that you have a year or more of hands-on cloud. If you are starting out, the Academy's Cloud Computing pathway is the right first step.
Is it an AI course?
It is an infrastructure programme. You will not train models. You will deploy, secure, monitor, cost and automate them on AWS and Azure, which is the part a cloud engineer is actually asked to own.
What do I need to have ready?
A reliable connection and a computer you can run a terminal on. The AWS sandbox and your n8n instance are provisioned for you, so you do not need your employer's accounts or your own card on a cloud provider.
What if I miss a Saturday?
Every session is recorded and available within 24 hours, and recordings stay open for 90 days after the programme. The lab and the Monday deliverable still apply, and the Slack community is where you catch up.
How does payment work?
Reserve above and we send payment details by email. Pay £799 in full, or £430 on enrolment and £430 before Week 3. Your place is confirmed when the first payment is received.
Can my employer pay, or send a team?
Yes. Two or more people from the same organisation get 10% off per head, and we can invoice the company directly. Choose "Corporate invoice" on the form.
Do I get a certificate?
On completion you receive the Cloudboosta AI-Native Cloud Engineer Certificate and a LinkedIn endorsement from Cloudboosta, after presenting your capstone in the final session.
Got questions? We're here