← Back to Teaching

Artificial Intelligence 1

Second semester 2026 · Foundations of Artificial Intelligence.

Description

The course offers a hands-on tour through data, classic machine learning, deep learning, computer vision, NLP, agents and MLOps. The goal isn’t to go deep on every technique, but for students to be able to frame a problem, build a baseline, compare alternatives, analyze errors and communicate a decision backed by evidence.

Taught over 14 sessions (2 holidays). Each session combines brief preparation, explanation, hands-on practice, and a compact individual deliverable.

Schedule

2026 edition, second semester.

Class Date Main topic
1 August 3 What is AI, the AI/ML/DL relationship, CRISP-DM and EDA on a real dataset
2 August 10 Modelable data, baseline, regression and MLflow. Case 1 launch
3 August 17 Classification, metrics and cost of error
4 August 24 Clustering, PCA and interpretation
5 August 31 Neural networks, MLPs and backpropagation
6 September 7 Optimization, regularization and evidence-based model selection
7 September 14 Midterm 1 · Introduction to CNNs, transfer learning and augmentation. Case 2 launch
8 September 21 Detection, segmentation and visual explainability
September 28 No class (Semana UCU)
9 October 5 Transformers, tokenization, embeddings and semantic search
October 12 Holiday — async activity: semantic retrieval
10 October 19 LLMs, prompting, RAG, grounding and evaluation
11 October 26 Agents, ReAct, LangGraph, MCP, tools, permissions and traces
November 2 Holiday — async activity: traces, permissions, guardrails
12 November 9 Midterm 2 · Introduction to MLOps and the end-to-end pipeline
13 November 16 Pipeline, release, rollback and final exam prep
14 November 23 Final exam: quiz, individual modification, group presentation and defense

Grading

Item Format Points
Five iRAT/tRAT Individual and team 10
Case 1 (portfolio) Team 7.5
Case 2 (portfolio) Team 7.5
Midterm 1 Individual, proctored 20
Midterm 2 Individual, proctored 20
Final exam Individual + group defense 35

Passing grade from 65%, with a minimum of 40% on the final exam. Teams of three students are the preferred format (pairs allowed when enrollment requires it).