AI Engineering

Build and operate AI systems.

Role-based pathways for engineers moving from product development into customer delivery, agent systems, model development, and production AI.

What this pathway developsThe ability to design, build, evaluate, and operate complete AI systems.

How to choose

Choose the part of an AI system you want to own.

Start from the role and responsibility, not the current tool. Each pathway builds a different part of the production AI lifecycle.

Programs

Explore each engineering pathway.

Open a programme to compare its scope, expected foundation, representative system, and practical outcome.

AI Engineer

Build production AI products end to end: LLM APIs, retrieval, agents and tool use, evaluation, guardrails, and deployment.

Open nowView course

Designed forSoftware engineers and technical builders moving into AI engineering roles.

Recommended foundationBasic programming experience. Web-development experience is useful but not required.

Product interfacesLLM APIsPrompt and context engineeringEmbeddings and retrievalAgents and tool useEvaluation

Practical outcome

A deployed AI product with retrieval, agents and tool use, evaluations, guardrails, and operational traces.

View current program

System mapA production AI system

Product interface

LLM + orchestration

Embeddings + retrieval

Agents + tools

Guardrails

Evaluation + observability

Deployment

Forward Deployed Engineer

Turn unclear customer problems into integrated, measurable technical solutions.

Open nowView course

Designed forEngineers moving into forward deployed, solutions, implementation, or technical consulting roles.

Recommended foundationIntermediate programming, API familiarity, and comfort explaining technical trade-offs.

Customer discoveryProblem framingData integrationRapid prototypesProduction deliveryObservability

Practical outcome

A customer-ready solution with integrated data, deployment evidence, and an adoption plan.

View course

System mapFrom customer problem to outcome

Customer problem

Discovery + scope

Customer data

Existing systems

Solution prototype

Integration + deployment

Adoption + outcome

AI Harness Engineer

Build the context, tools, memory, evaluation, and controls that make model behavior useful and reliable.

Coming soonLearn more

Designed forBackend, platform, and AI engineers building agentic applications around foundation models.

Recommended foundationComfort with Python or TypeScript, APIs, data structures, and backend service design.

Context constructionRoutingRetrievalTool contractsMemoryEvaluations

Practical outcome

A controlled agent harness with retrieval, tools, persistent state, approvals, evaluations, and traces.

Explore planned curriculum

System mapA reliable agent harness

User goal

Context + router

Retrieval + memory

Tools

Model

Evals + guardrails

Tracing + improvement

ML & Deep Learning Engineer

Develop data, training, evaluation, and serving workflows for machine-learning and deep-learning models.

Coming soonLearn more

Designed forSoftware, data, and analytics professionals moving into model development.

Recommended foundationPython, basic statistics, linear algebra fundamentals, and confidence working with tabular data.

Data preparationFeature pipelinesTrainingExperiment trackingModel evaluationRegistries

Practical outcome

A reproducible model pipeline with training, experiment tracking, evaluation, deployment, and monitoring.

Explore planned curriculum

System mapA model development lifecycle

Training data

Experiments

Feature pipeline

Model training

Evaluation

Model registry

Serving + monitoring

LLM Engineer

Prepare datasets, adapt language models, evaluate behavior, and optimize inference.

Coming soonLearn more

Designed forML and AI engineers specializing in language-model development and adaptation.

Recommended foundationPython, PyTorch fundamentals, machine-learning evaluation, and familiarity with transformer models.

Dataset preparationTokenizationSFT and LoRAPreference tuning and RLVRBehavior evaluationInference

Practical outcome

An adapted language model with documented data, reproducible training, behavior evaluations, and an inference endpoint.

Explore planned curriculum

System mapAn LLM adaptation workflow

Training dataset

Data preparation

SFT + LoRA

Preference tuning

Model evals

Inference optimization

Product integration

LLMOps Engineer

Deploy and operate language models with reliable serving, telemetry, scaling, security, and rollback.

Coming soonLearn more

Designed forPlatform, DevOps, MLOps, and AI engineers responsible for production model infrastructure.

Recommended foundationLinux, containers, networking, cloud infrastructure, and basic model-inference concepts.

Model registryRelease pipelinesInference gatewaysGPU servingAutoscalingTelemetry

Practical outcome

A production model platform with controlled releases, autoscaling, telemetry, security, and rollback.

Explore planned curriculum

System mapA production model platform

Model registry

Release pipeline

Inference gateway

Model serving

GPU capacity

Metrics + traces

Scale + rollback

Across every pathway

Learn the system through working software.

01Understand the system

Learn the interfaces, data, models, controls, and infrastructure behind the capability.

02Build working components

Turn concepts into services, integrations, evaluations, and deployable systems.

03Develop engineering judgment

Make trade-offs, investigate failures, and explain why the system behaves as it does.