Self-paced
The full recorded library, the browser GPU lab and every project brief — on your own schedule. Add mentor reviews whenever you want them.
Every course is part-time and mentor-reviewed, runs in your browser on our GPUs — one click, nothing to install — and is built around real project work with a capstone tied to a live role. Filter by track, then apply to the cohort that fits your goal.
Every track is available in four delivery formats. Same curriculum, same one-click GPU lab, same capstone — you choose the amount of live contact, the schedule and the setting that fit your life. Pricing scales with the format.
The full recorded library, the browser GPU lab and every project brief — on your own schedule. Add mentor reviews whenever you want them.
Scheduled virtual classes with real-time Q&A, the full one-mentor-to-eight code review, and a cohort keeping pace with you. Fully remote.
Recorded core lessons through the week, then live weekend sessions for labs, review and project work. Built around a full-time job.
Classroom cohorts at our centre — in-room instruction, on-site mentor time and a study group, plus everything in the live online format.
You can usually switch or upgrade format between modules.
Prices vary by track. Instalment plans and early-bird pricing available — confirmed on your advisor call. Next cohort starts [date]. Self-paced enrolment is always open.
12 courses shown
NumPy, pandas, linear algebra, calculus and probability — the working toolkit behind every model, taught through code.
Supervised and unsupervised learning, model evaluation, and the scikit-learn workflow. Train, tune and explain your first models.
Tensors, autograd, training loops, and the core architectures. Build MLPs and CNNs from scratch, then with PyTorch.
Prompt design, retrieval-augmented generation, tool use, agents, guardrails, evaluation and cost control — ship an LLM product end to end.
Chunking, embeddings, hybrid search, re-ranking and evaluation. Build a retrieval layer that holds up under real queries.
LoRA/QLoRA, instruction tuning, preference data, and building evaluation suites you can trust before you ship.
Batch and streaming pipelines, orchestration, data quality and feature stores — get clean, reliable data to your models.
Containerise, deploy, monitor and retrain models with CI/CD, model registries and observability you can reuse at work.
Serving frameworks, batching, caching, quantisation, cost/latency trade-offs and monitoring for LLM workloads.
Classification, detection and segmentation with modern backbones. Data pipelines, augmentation, transfer learning and deployment.
Tokenisation, attention, pretraining and fine-tuning for classification, extraction and sequence tasks with Hugging Face.
Combine image and text models: captioning, visual question answering, document understanding and OCR-driven pipelines.
Book a short call with an advisor. We'll look at your background and goals and recommend a starting point.
Talk to an advisor