Revolution of Mind

By Rohan Manzoor

Nine weeks · One continuous argument

Learn how intelligence became buildable.

The course begins with the structures that make learning possible and ends with generative systems that can retrieve, reason, and act.

See the complete journey
01

Understanding is not collected. It is constructed.

Foundation

See the structure.

Geometry, calculus, probability, and computation become working ideas rather than prerequisites to survive.

Practice

Make it move.

Every concept crosses into code, experiment, diagnosis, and an engineering decision.

Achievement

Build what comes next.

The final weeks connect models to deployed systems, transformers, retrieval, and agents.

The curriculum

From the language of nature to machines that generate.

01

Mathematical language

Vectors describe possibility, calculus describes change, and probability describes belief.

  • Vector geometry
  • Calculus & gradients
  • Probability & Bayes
P(θ∣x)=P(x∣θ)P(θ)P(x)P(\theta \mid x)=\frac{P(x\mid\theta)P(\theta)}{P(x)}
Lectures

02

Computing foundations

A model is only useful when ideas can be expressed as data, operations, and reliable programs.

  • Python
  • Data structures
  • Algorithms
f:X→Yf: X \rightarrow Y
Week

03

Seeing data

Before prediction comes perception: structure, quality, scale, and the stories data can and cannot support.

  • EDA
  • Data quality
  • Visual reasoning
zi=xi−μσz_i=\frac{x_i-\mu}{\sigma}
Week

04

Statistical reasoning

Statistics turns variation into evidence—provided we make assumptions visible and test them.

  • Distributions
  • Inference
  • Experimental thinking
xˉ±z∗σn\bar{x} \pm z^*\frac{\sigma}{\sqrt{n}}
Week

05

Learning from error

Learning begins when error becomes measurable, navigable, and connected to a decision.

  • Regression
  • Loss
  • Optimisation
  • Metrics
θt+1=θt−η∇θL(θt)\theta_{t+1}=\theta_t-\eta\nabla_\theta \mathcal{L}(\theta_t)
Week

06

AI systems

A model becomes engineering when it survives interfaces, change, failure, and observation.

  • APIs
  • CI/CD
  • Deployment
  • Monitoring
system=model+data+feedback\text{system} = \text{model} + \text{data} + \text{feedback}
Week

07

Decision geometry

Trees partition, neighbours compare, and margins separate—three views of the same act: deciding.

  • Decision trees
  • KNN
  • SVM
y^=sign⁡(wTx+b)\hat{y}=\operatorname{sign}(w^Tx+b)
Week

08

Neural networks

Representation emerges through layers; memory emerges when a network can carry state through time.

  • Neural networks
  • RNN
  • LSTM
  • Attention
ht=tanh⁡(Wxxt+Whht−1+b)h_t=\tanh(W_xx_t+W_hh_{t-1}+b)
Week

09

Generative intelligence

Transformers turn context into a field of relationships—and generation into repeated prediction.

  • Transformers
  • Fine-tuning
  • RAG
  • Agents
Attention⁡(Q,K,V)=softmax⁡ ⁣(QKTdk)V\operatorname{Attention}(Q,K,V)=\operatorname{softmax}\!\left(\frac{QK^T}{\sqrt{d_k}}\right)V
Week

Lectures come next

The curriculum is mapped.
The stage will be built deliberately.

See the lecture direction