What do I need to know before starting?

The entry expectations are listed on this page. If you are close but not quite there, tell us at application; we run short pre-course preparation material and would rather get you ready than turn you away.

Does this course include the internship?

Yes, for learners who complete the core modules and meet the mid-point assessment bar. The internship is on live client work with a supervisor and code review, and you receive a written reference describing what you actually did.

Is there a certificate?

Yes, on completion of the capstone. More usefully, you finish with deployed projects, a public repository history, and a supervisor’s reference. Those are what interviews actually turn on.

What roles does Deep Learning prepare me for?

Typical destinations are Deep Learning Engineer, Machine Learning Engineer and AI Developer. The curriculum is built backwards from what those interviews and first ninety days actually require.

Academy track60 hours · Intermediate

Deep Learning

Name: Deep Learning
Availability: InStock

ANN, CNN, sequence models and deployment

Apply for this trackPreview the syllabus

Fee on request

Duration: 60 guided hours
Level: Intermediate

Apply now

Next cohort intake is open · instalments available

Why this track

How neural networks actually learn, and how to train, evaluate and ship them. You build fully connected networks, CNNs for image classification and sequence models, reuse pretrained weights through transfer learning, and deploy a trained model behind a Streamlit interface.

60hours of guided, hands-on learning

7modules, each ending in something deployed

3portfolio projects, reviewed like client work

3+roles this track prepares you for

Outcomes

What you will be able to do

Explain neural-network training and common optimization choices
Implement ANN, CNN and sequence-model examples
Evaluate overfitting, compare models and use pretrained weights
Deploy a trained deep-learning model through a Streamlit interface

Before you start

Entry expectations

If you are close but not quite there, tell us at application. We would rather get you ready than turn you away.

01Python and NumPy
02Basic ML concepts (the Machine Learning course is recommended first)
03Elementary linear algebra

Tools you will use

PythonNumPyPyTorchTensorFlow/KerasStreamlitGit & GitHub

Syllabus

A first look at the route

7 modules over 60 guided hours. Here is how it starts; the full roadmap opens up when you join.

M01Neural network foundations
- Perceptron, layers, weights, bias and forward propagation
- Activations: sigmoid, tanh, ReLU, softmax
- Loss functions, gradient descent and backpropagation intuition
- Overfitting, validation and learning curves
M02Framework and training workflow
- Tensor operations, datasets, dataloaders and batches
- Optimizers, learning rates, epochs and checkpoints
- GPU/CPU selection, reproducibility and experiment logging
- TensorFlow/Keras comparison and model-saving formats

5 more modules in the full roadmap

The complete syllabus, with every topic, hands-on lab and the capstone brief, is shared with learners when they join the course.

Join the course

Portfolio

What you will have built

01An image or sequence learning application with dataset preparation, baseline, training history, evaluation and a saved model
02Streamlit demos for an ANN, a CNN image classifier and a sequence model
03A documented GitHub repository with a Streamlit inference UI

Where it leads

Roles this prepares you for

Deep Learning EngineerMachine Learning EngineerAI Developer

Questions

Before you apply

The course is 60 guided hours of explanation, guided coding and review, plus independent practice; admissions will share the cohort schedule. Live sessions are scheduled outside standard office hours and recorded, so working professionals can keep a full-time job while studying.