AI Engineering Camp
HANDS-ON · 10 SESSIONS · 30 HOURS
From understanding AI models to building demo-ready, deployable AI products.
An advanced, hands-on training camp for final-year students and graduates of technical majors. It combines model building, AI systems engineering, teamwork, and the development of a final project. Its goal is to take participants from academic knowledge and isolated experiments in Jupyter Notebooks to building complete solutions with practical value, using ways of working close to real product development environments.
Camp methodology
- Understand
- Apply
- Build the project
Prerequisites
The camp is not for complete beginners. Familiarity with Python, programming basics and machine learning concepts is recommended, along with the ability to learn and practice independently between sessions.
Who it is for
- Final-year students in informatics engineering, telecommunications engineering and related majors.
- Graduates aiming to work in AI engineering.
- Developers who know Python and ML/DL basics and want to move on to building practical systems and products.
Learning path
Four progressive stages · ten sessions · thirty hours
Foundations
Session 1Engineering tools, collaboration and code management.
Models
Sessions 2–5Machine learning, deep learning, computer vision, natural language processing.
Systems
Sessions 6–9LLM engineering, AI-assisted coding, agents, voice and multimodal systems.
Product
Session 10Deployment, evaluation and the final demo.
The ten sessions
| # | Session | Key practical outcomes |
|---|---|---|
| 01 | Git, GitHub & AI Engineer Workflow | Set up the environment, manage branches, open a first Pull Request and resolve a merge conflict. |
| 02 | Machine Learning: Fundamentals → Product | An end-to-end prediction model with FastAPI, Streamlit and Docker. |
| 03 | Deep Learning: From Scratch → PyTorch | A neural network from scratch, then a PyTorch application with training curves and performance analysis. |
| 04 | Computer Vision: Classical → Detection & Tracking | An image classification model and a hands-on object detection and video tracking exercise. |
| 05 | NLP & Transformers | Fine-tune a BERT model for named entity recognition (NER) and evaluate it. |
| 06 | LLM Engineering: APIs, Prompting & RAG | A RAG assistant over the camp’s documents with a user interface. |
| 07 | AI-Assisted Coding: Copilot, Claude Code & Codex | Develop a feature with an agent and submit a PR with review and tests. |
| 08 | Agentic AI: Tools, MCP & Multi-Agent Systems | An agent that uses tools and an MCP server, with traces of successes and failures. |
| 09 | Voice & Multimodal Agents with Pipecat | An interactive voice assistant, with visual input where possible. |
| 10 | Production, Evaluation & Demo Day | CI/CD, deployment and monitoring basics, then the final project demos. |
Capstone Project
The core idea
Participants work in teams to build a runnable, demo-ready AI product that addresses a clear problem and demonstrates design, implementation, evaluation and documentation skills.
Suggested project ideas
- A source-cited research assistant
- A RAG-based study assistant
- A data extraction system for Arabic documents
- A voice-based English learning assistant
- A computer vision application
Design principle
RAG, agents or voice AI are not required in every project. The architecture must start from the needs of the problem, not from a wish to stack as many technologies as possible.
Project milestones
- End of session 06Form teams, pick a problem, and submit a short proposal.
- Session 07Design the solution and start development with AI coding tools.
- Session 08Integrate AI capabilities, tools or agents where needed.
- Session 09Finish integration and experiments; add voice or multimodal input if it suits the project.
- Session 10 · Demo DayTesting, the final demo, and a discussion of the engineering decisions.
Training and evaluation
Learning by building and reviewing, as in real work environments.
Focused explanation
The essential concepts and engineering decisions.
Live Coding
Building a core part of the project in front of participants.
Pair Lab
Group practice to solve a problem or complete a feature.
Homework
Assignments submitted as Pull Requests to the camp repository.
Code Review
Practical feedback on the quality of the implementation.
Demo Day
The final project is evaluated on idea, implementation, quality of results, documentation and presentation.
Typical structure of a 3-hour session
The rhythm is adjusted for shorter sessions.
Final project evaluation criteria
- Idea
- Implementation
- Quality of results
- Documentation
- Presentation
Outcomes and hiring
By the end of the camp, participants are expected to be able to
- Choose the right model or technical approach for a problem, then train and evaluate it.
- Turn models into services and applications using APIs and user interfaces.
- Build applications based on LLMs, RAG and AI agents.
- Use AI-assisted coding tools effectively and responsibly, with code review and testing.
- Collaborate using Git, GitHub and Pull Requests.
- Apply the basics of deployment, monitoring, evaluation and security.
- Deliver a documented, runnable technical project that strengthens their professional profile.
Expected deliverables
- 1A GitHub repository with the assignments and practical applications.
- 2Hands-on experience building models and turning them into services.
- 3Practical applications in RAG, agents and voice AI.
- 4Experience writing requirements, using coding agents, and reviewing their output.
- 5A documented capstone project with setup instructions, evaluation results and a demo.
- 6Technical material to showcase in a GitHub portfolio and on LinkedIn.
Referrals to companies with a personal recommendation
We have coordinated with a number of companies to refer camp graduates for employment.
Trainers
The camp is led by eight engineers and experts working at local and international companies. Meet them on the team page.
Join our community
Follow our activities, workshops and camps, and share your experience on social media.