Track overview
AI & MLOps Track
Outline the AI and MLOps track's focus on applied AI, LLM application patterns, and deployment workflows so learners can decide if it matches their AI career direction.
Focus
Applied AI and LLM application patterns
Emphasis
Model deployment and production workflows
Format
Hands-on modules with guided builds
What this track is built around
The AI and MLOps track is for learners who want to work with machine learning models after they are trained — turning prototypes into services that other systems can rely on.
Applied AI
You work with datasets, features, and evaluation routines that measure whether a model behaves well on real inputs. The point is judgement — knowing what a model is doing, where it fails, and what to change next.
LLM patterns
You build LLM application patterns — retrieval, structured prompting, tool use, and guardrails — and learn how each pattern changes cost, latency, and reliability.
Deployment
You package models, define services around them, and wire deployment workflows so a model can be shipped, versioned, and rolled back without guesswork.
Operations
You watch live services: tracking inputs and outputs, detecting drift, and setting up alerts that tell you when a model needs attention before users notice.
Where the track sits in your learning path
This is one of the academy's structured pathways. Compare it against the others before you commit.
Systems & defence
Cybersecurity Track
Network defence, systems hardening, and analyst workflows for learners heading toward security operations roles.
View cybersecurity trackCurrent track
AI & MLOps Track
Applied AI, LLM application patterns, and the deployment workflows that keep models running in production.
See module outlineBuild & ship
DevOps & Cloud Track
Pipelines, containers, and cloud infrastructure — the operating layer that AI services eventually run on.
View DevOps trackModule outline
The track runs as a sequence of hands-on modules. Each one ends with something you build and can talk about, rather than a quiz score.
Foundations
Python, data handling, and model basics
Refresh the parts of Python that matter for data work, then move through loading, cleaning, and splitting a dataset. You train a small model, inspect its outputs, and learn to read the difference between a model that is wrong and a model that is unlucky.
Applied AI
Evaluation, features, and honest measurement
This module is about deciding whether a model is actually useful. You build evaluation sets, choose metrics that match the problem, and practise explaining a result to someone who has to make a decision from it. Feature work sits here too — what to keep, what to drop, and why leakage ruins a good score.
LLM applications
Retrieval, prompting, and tool use
You build LLM application patterns one at a time: retrieval over your own documents, structured prompting with a defined output shape, and calling tools from inside a model response. Each pattern is compared on cost, latency, and how it fails, so you learn when a simpler approach is the better answer.
Deployment
Packaging models and shipping a service
You take a trained model, package it with its dependencies, and expose it behind a service with a clear input and output contract. Then you deploy it, watch it under load, and practise releasing a new version without taking the old one down.
Operations
Monitoring, drift, and rollback
The final module covers what happens after launch. You log inputs and outputs, track the metrics that hint at drift, and set alert thresholds you can defend. You also rehearse a rollback, because the skill is knowing what to do when a deployed model stops behaving as expected.
Hands-on labs
Labs that mirror a real deployment
Lab work in this track happens in environments that look like the systems you would meet on a team: a place to run training jobs, a place to serve a model, and logs to read afterwards. You practise the full loop — change something, ship it, watch what happens, and decide whether to keep it.
Each lab is scoped so you can finish it in one sitting and explain your choices in a short written summary. Those summaries are what you keep — they become the record of what you built and why.
Skills you should leave with
These are the working habits the track is designed to build. They are the things you would describe in an interview when asked what you can actually do.
Evaluate
Measure before you tune
Build a test set you trust, pick metrics that match the decision, and avoid chasing a number that does not reflect real use.
Design
Choose the right pattern
Decide between retrieval, prompting, fine-tuning, or a plain rule — and defend the choice with cost, latency, and failure behaviour.
Ship
Deploy and version models
Package a model, define its service contract, and release updates in a way that lets you step back safely if something breaks.
Operate
Watch it in production
Track inputs, outputs, and drift indicators, then set alerts that give you time to react instead of a page at midnight.
How the format works
The track is delivered online, so you can work through modules around a job or other study. Nothing here runs on a fixed daily schedule.
Self-paced
Modules open in sequence so the foundations land before the deployment work. You set the weekly pace.
Guided builds
Each module pairs written explanation with a build task. You are expected to type the code, run it, and fix what breaks.
Review
Written summaries of your lab work give you a record of decisions and reasoning you can return to later.
Prerequisites
Comfortable with basic programming and ready to read technical documentation. No prior production experience is assumed.
Who this fits
Good fit, and honest limits
The track suits people who already enjoy building things and want to move toward AI developer work — whether you are a developer adding machine learning to your toolkit, a data analyst moving closer to production systems, or a graduate building a portfolio from scratch.
It is not a research programme. If your goal is publishing papers or designing novel model architectures, this track will feel practical rather than theoretical. It also assumes you will put in the hours at a keyboard; watching the material alone will not build the skills.
Completion leads to a track record of finished lab work, not a job offer. Hiring decisions stay with employers, and outcomes vary by background, effort, and market conditions.
Questions learners ask
Short answers to the things people usually want to know before starting this track.
Do I need machine learning experience before starting?
No. The foundations module starts from Python and data handling, then introduces model training in small steps. What helps more than prior machine learning is being comfortable reading documentation and debugging your own code.
Is the LLM work about building models or using them?
About using them well. You work on LLM application patterns — retrieval, structured prompting, and tool use — and learn how those choices affect cost, latency, and reliability. Training large models from scratch is outside the scope of this track.
How much time should I plan for each week?
Plan for several focused hours a week if you want the lab work to stick. Because the track is self-paced, the exact number depends on how much time you spend debugging and writing up what you built.
Does this track cover deployment to cloud platforms?
Yes, deployment workflows are a core part of the track. If infrastructure itself is your main interest rather than the model, the DevOps and cloud track goes deeper on pipelines, containers, and cloud environments.
What should I do if I am unsure this is the right track?
Read the module outline above and compare it with the cybersecurity and DevOps tracks. If you still cannot decide, get in touch and describe where you are now — we can talk through which pathway matches the work you want to be doing.
Ready to look closer?
If the AI and MLOps track matches the direction you want, the next step is a conversation about your background and what you want to build. We can also point you at the track that fits better if this one does not.
Visit
452 Tech Plaza, Suite 300, San Francisco, CA 94105, USA
Call
+1-415-555-0192
Hours
Monday to Friday, 9:00 AM – 6:00 PM