# AI Project Gallery: an index of thirty-three student machine learning projects > A repository containing one file, which is a table. That turns out to be a more useful artifact than most portfolio repositories, because it is browsable. **KalyanM45/AI-Project-Gallery** — This Repository Contain All the Artificial Intelligence Projects such as Machine Learning, Deep Learning and Generative AI that I have done while understanding Advanced Techniques & Concepts. - Repository: https://github.com/KalyanM45/AI-Project-Gallery - Stars: 6,820 · Forks: 1,304 - Language: Unknown - License: not declared - Published: 2026-10-07 · Updated: 2026-10-07 · Language: en - Canonical page: https://hysenlabs.com/projects/kalyanm45-ai-project-gallery ## One file, thirty-three rows, four columns The entire repository is a single `README.md`. There is no code, no license, no contributing guide and no build. That is unusual for a repository with 6820 stars and it is the point rather than a shortcoming. The main table has five columns: serial number, project name, domain, a repository link rendered as a GitHub Repo badge, and an end-to-end marker. The serial numbers run from 01 to 33, which means the index has grown over time rather than being written all at once. The domain column is the most informative part, because it is the only editorial judgement in the whole document. Projects are labelled Classification, Regression, Web Scraping, Generative AI, MS Power BI, Computer Vision, Recommendation System, Deep Learning, Data Analytics, Agentic Workflows or Agentic AI. A reader can filter for a technique without opening a single link. ## What the domain distribution says about the field Counting the labels gives a picture of what a self-taught machine learning portfolio looks like in practice. Classification is the largest group by a wide margin: Airbnb price prediction, chest disease detection, diamond and gold price prediction, diabetes, heart disease, password strength, rock versus mine, spam email and wine quality all sit under it. Regression covers the Boston housing set, flight fares and student performance. Web scraping has exactly two entries, an article scraper and an image scraper, both end-to-end. Computer vision has one, hand tracking with OpenCV. The recent entries tell a different story from the early ones. The first twenty-seven rows are the classic curriculum: tabular datasets, a recommendation system, a few chatbots. Rows 28 through 33 are agent-shaped: Market Insight tagged Agentic Workflows, a travel planning agent tagged Agentic AI, a multi-agentic blog generation project, and a GitHub tracker. Two of those use the author's second account name rather than the one the older projects use, which suggests the newer work is being done separately. ## The end-to-end column is the only quality signal The checkmark column records which projects run end to end, and it is not uniformly filled. Among the early entries, Airbnb price prediction, Boston housing, chest disease, diamond price, flight fare, heart disease, movie recommendation, spam email and student performance are marked. Article scraping, the Gemini and OpenAI chatbots, diabetes prediction, hand tracking, the medical assistant and wine quality are not. That is a small but real signal, and it is the kind of thing most portfolio repositories omit entirely. A notebook that trains a model and prints a metric is different from a project with data loading, training and inference wired together, and the author is distinguishing between them in public. Whether the unmarked projects are incomplete or simply not packaged as end-to-end apps is not something the index can tell you. You would have to open them. ## A second table for projects that do not exist yet Below the main table, under a heading announcing that many more projects will be uploaded soon, there is a second table with the same columns minus the end-to-end marker. Its repository cells contain Coming Soon badges instead of links. The visible entries are deep fake detection and arrhythmia disease detection. So the gallery is honest about what it does not have yet, which is a small courtesy that costs nothing. This second table also shows how the index will grow. It is the same structure with the end-to-end column dropped, which suggests the author considers a project without that claim not yet worth tracking. ## What this repository is and is not It is a portfolio index and a discovery list. It is not a library, not a course, and not a collection of code you can depend on. The description frames it as projects done while understanding advanced techniques and concepts, which is the right framing for this kind of repository. Topics reinforce that intent, listing fourteen tags that are almost entirely about the projects rather than the repository: machine learning projects for beginners, deep learning projects, generative AI projects, computer vision projects, NLP projects, LLM projects, data science projects, and Power BI projects. Open issues sit at two and forks at 1304, which is a fork rate well above what a code library would draw. Forking an index of links is a reasonable thing to do if you want your own curriculum track, and it is also a reminder that the stars here reflect the author's visibility rather than the code's reuse. The repository was pushed to on 2026-08-26. No license is declared, which for an index of links is defensible but means the table itself has no explicit terms. ## Conclusion This repository does exactly one thing and does not obscure it: it is a table. Thirty-three projects, sorted by serial number, tagged by domain, linked to separate repositories, with a checkmark column recording which ones are end-to-end. That structure makes it useful in a way a pile of project folders never is, since you can scan for the domain you care about rather than reading thirty READMEs to find out. The limits are equally clear. Every project lives elsewhere, so the gallery tells you nothing about quality, and it will drift out of date the moment a linked repository changes. Use it the way the author intends, as an index to jump from, then read the individual repository before you judge the work. ## FAQ ### What is in the AI Project Gallery repository? A single README file containing a table of thirty-three projects, each with a serial number, a project name, a domain label and a link to a separate repository. There is also a second table listing planned projects that are marked as coming soon. No code lives in this repository itself. ### How do I find a project in a specific area, like NLP? Use the Domain column in the main table. Each row is tagged with one of a fixed set of labels including Classification, Regression, Web Scraping, Generative AI, Computer Vision, Recommendation System, Deep Learning, Agentic AI and MS Power BI, so you can scan for the technique you want before opening any link. ### What does the end-to-end column mean? It marks which projects run as complete pipelines rather than isolated notebooks. Projects with a checkmark in that column have been packaged to cover more than just model training. Several entries are left unmarked, including some of the earlier classification projects. ### Does the gallery include generative AI and agent projects? Yes. There are several generative AI entries built on Gemini and OpenAI models, and the four most recently added rows are agent-shaped: an agentic workflow project for market insight, a multi-agentic blog generation tool, a GitHub tracker and a multi-agent travel planner. ## Sources - [Issues](https://github.com/KalyanM45/AI-Project-Gallery/issues) - [KalyanM45/AI-Project-Gallery on GitHub](https://github.com/KalyanM45/AI-Project-Gallery) - [README](https://github.com/KalyanM45/AI-Project-Gallery/blob/main/README.md) --- Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/kalyanm45-ai-project-gallery