▲ Project Ideas

Don't feel limited to this list. These are just a few starting points. Get creative! The applications are limitless.

Looking for a GitHub repo template to jump in and get started? Look no further! Hack away the day and clone yourself a HyperTech Workshop 🚀 (opens in a new tab).

GL, HF!

Find New GitHub Stars

A handful of projects from my Github Stars ⭐ (opens in a new tab)!

Start a Virtual Company w/ ChatDev 👩🏾‍💻👨🏽‍💻👩🏻‍💻 (opens in a new tab)

Run a Simulation w/ Generative Agents 🌿🧑🏾‍🤝‍🧑🏽🧑🏽‍🌾 (opens in a new tab)

Click Here For More! (opens in a new tab)

Explore Popular LLM Tools

LangChain: Build context-aware, reasoning applications with LangChain’s flexible abstractions and AI-first toolkit.

LlamaIndex: A simple, flexible data framework for connecting custom data sources to large language models.

Chroma: Chroma is the open-source embedding database. Chroma makes it easy to build LLM apps by making knowledge, facts, and skills pluggable for LLMs.

LangChain 🦜

LlamaIndex 🦙

Chroma 🟡

Retrieval Augmented Generation

Follow any one of the videos or workflows below to dive into retrieval augmented generation (RAG). With RAG, you can give your chatbot or language model long term memory and extended context lengths with abilities to retrieve and augment responses based on an index or vectorbase you build from your content.

James Briggs on Youtube 📺 (opens in a new tab)

NVIDIA NeMo

Follow any one of the videos or workflows below to dive into NVIDIA's new colang language and NeMo Guardrails library. This unique approach is worth your time if you need to align your model with specific functions and interactions. Keep an eye on colang.

NVIDIA Resources 🖥️ (opens in a new tab)

James Briggs on Youtube 📺 (opens in a new tab)

Cog: Containers for Machine Learning

Replicate Resources ⚙️ (opens in a new tab)

Self-Development

Learn Faster w/ AI Tutors & Teaching Assistants

Design and chat with your own AI tutor to help understand the contents of this site.

1.) Draft a prompt for a personal tutor / assistant persona.
2.) Choose a model accessible to you (GPT-3.5, Open-Source, etc.) and get it running on a text-based platform of choice.
3.) One-Shot prompt the model with your prompt and see how it responds.
4.) Begin asking it questions pertinent to your topic or goal.
5.) Start copy/pasting contents from this site into the model and ask it to teach you w/ custom methods or explanations.

Organize Yourself w/ AI Interns & Project Managers

One AI not enough? Design and chat with an entire staff of AI assistants, interns, and project managers to help organize your thoughts, projects, or notes.

1.) Draft a prompt for a project manager and team of assistants.
2.) Choose a model accessible to you (GPT-3.5, Open-Source, etc.) and get it running on a text-based platform of choice.
3.) One-Shot prompt the model with your prompt and see how it responds.
4.) Begin asking it questions pertinent to your topic or goal, try asking members of your group or team and see if the model picks up the multiple personas.
5.) Organize your digital files, notes, or goals with the assistance of the project manager and team. Combine this prompt with the your personal learning assistant to multiply how many angles and perspectives you can infer with at a given time.

Chat w/ Notes, Articles, Files, and Documentation

Feed your text-based content to a LLM and 'chat' with the contents of your file or resource.

1.) Find an open-source chat document Q&A program from my GitHub Stars (opens in a new tab) and see if you can get a platform installed and running.
2.) Choose a model accessible to you (GPT-3.5, Open-Source, etc.)
3.) Upload your document(s) to the software platform or build your own pipeline w/ something like LangChain (opens in a new tab) or Llamaindex (opens in a new tab).
4.) Begin asking your document questions pertinent to your topic or research goal!
5.) Upload more documents and reindex vectors as needed.

✍️ Contribute to FOSAI ▲ XYZ

Want to place your project idea here? Contribute to this wiki by adding your project or idea to this page!

projects.mdx or demos.mdx

First, clone the repo (opens in a new tab) to your device and then run pnpm i in your terminal of choice to install the dependencies.

Then, run pnpm dev to start the development server and visit localhost:3000.

From here, you should be able to see the 'pages' folder, which contains all of the webpage content you see here (editable in simple markdown).

Demos Prompts