👋🏻 Hi there! This is the cohort repo for Special Topics: AI Coding Tools – Tech Talent Pipeline, Winter of 2026 at BMCC. This is your single source of truth for all materials related to the course, namely assignments, recorded lectures, slides, and code demos.
We'll be exploring the use of Large Language Models (LLMs) to augment your software engineering process. The following themes will recur throughout this course:
⚠️ Responsibility. You are responsible for the code you submit, whether or not it was written by an LLM. You will be prompting tools like Claude and Gemini to write code on your behalf, but they will interpret your instructions in ways you do not anticipate. Sometimes they will hallucinate. It is your duty to review what they have written and ensure that you are not adding more problems than you are solving, including security vulnerabilities. You are also responsible for your own learning development, and be mindful that you aren't letting these generative tools stunt your growth. We will also discuss the broader social impact of AI and the ethical responsibility you bear as a worker in the tech industry.- 🔬 Testing. Automated testing is an integral part of pretty much every organization that writes software. As a project grows in scale, it becomes increasingly difficult to manually verify that existing functionality hasn't broken. Automated tests allow engineers to perform this verification automatically by clearly specifying the expectations of the code. With LLMs in the picture, they gain a new utility: guiding the agent to generate code that conforms to the specifications expressed in these automated tests. Getting good at testing will help you become productive with AI Coding Tools.
- 🏭 Large Codebases. Generative AI is pretty good at prototyping simple apps from the ground up. It's already seeing a lot of usage in the tech industry as "vibe coding." A more challenging (but more useful) application of LLMs involves working within an existing codebase. You'll spend time exploring a real open-source project and use Chat to speed up your familiarity with the codebase. This will also give you a chance to contribute to an open-source application with real users!
This course runs from January 2 through January 21. You will be asked to create Pro accounts with several AI providers, such as Anthropic and Cursor. These services have one-month free trials, and I suggest you start such a trial and cancel it before the trial expires.
- 📺 Large Language Models explained briefly
- 📖 What is Tokenization - Hugging Face
- We won't actually be building our own LLMs in this course, but you should skim this blog post to get a sense for how these models are engineered. Hugging Face is a good learning resource for this sort of thing.
- Also, check out the 🕹️ Tokenizer Playground linked in that blog post.
- 📖 Exclusive: OpenAI Used Kenyan Workers on Less Than $2 Per Hour to Make ChatGPT Less Toxic
- 📖 The Emerging Problem of "AI Psychosis"
- 📖 Linux Kernel’s AI Code Revolution: Guidelines for the Machine Age
Project #1 is due at 11:59pm on January 8
😎 No lecture today, just breakout rooms and exploring fun open-source projects to work on.
Enter which open source project you want to work on for Project #2 by 11:59pm on January 13