Toronto Accelerationists
Presentation
New Description
There is an old Chinese saying, "talk does not grow rice". It is easier to talk about AI than build something that integrate into people's lives. Why talk about what can be when you can be the change you want to see in the world.
Change requires knowledge, skills, and coordination, with all this information at our fingertips knowing what to ignore is more important than knowing what one should know.
So let's get our values articulated so we can converse and build a future worth getting excited about.
OG Description
Join the Discord: https://discord.gg/j9DVjpguwd
A community for builders to come together, share what they are building, get feedback, learn new skills, and explore the ideas that will shape the next 1000 years of humanity.
e/acc or Effective Accelerationisim started as a meme on twitter that has not evolved into an online subculture that promotes building, technocapital, and the preservation of consciousness as a force within the universe.
We have both online and in person meetups tbd
Ai meetup fixes
- Get friends and Kromar to do tutorial
- Research YouTube videos
- Epic example demo people can use
Langchain
- Have a frontend for people to play with embeddings
- Have a syncing functionality to share sets of embeddings as collections
-
Do the langchain tutorial on the browser
Discussion Group
- What is Accelerationism?
- What AI sci fi future do you want to come true?
Notes
- What is this space?
- What is this Group?
- e/acc Effective Accelerationisim is a reaction to effective altruism
- Let's walk confidently into the future rather than trying to regulate everything because we are scared. AI may have growing pains, but so did the industrial revolution.
- Accelerationisim encourages Building, Using and Testing rather than regulating
- I want to build stuff and to do that I need to learn
- The best way to learn is by teaching
- I want to inspire people to build stuff and in return have them inspire me
- What are we building?
- Build Agents to read and talk about and compare text such as documents, books, and even social media information
- Introductions
- So Accelerationisim
- What future do we want?
- What do we want to do once we have ASI(Artificial Super Intelligence)?
- Who here wants to live in a future like Star Trek?
- If AGI is achieved what role should AI have in raising our children?
- Should we give people basic income or a humanoid robot that can make money for them?
- Do you want to live in the matrix but have control over your own simulation?
- How do we avoid humanity becoming a parasite on the new sudo intelligent systems that will inherit the earth? Parasites survive off other organisms and therefore don't need complex organs like eyes, brains, and ears that are used to survive in a complex environment.
- I encourage those interested to stay after the workshop to discuss this over some tea.
- Awesome Lists
- Merlin AI
- OpenAI Integrated in every web page
- Twitter Integration
- Create your own ChatBot (System Prompt)
- I am canceling my ChatGPT subscription and using Merlin
- Where are we Going?
- LangChain on X: "💬"LangChain Tool Calling feature just changed everything" We recently added a common interface for tool calling across model providers This makes it easy to build agents that work across models Watch @EdenEmarco177 explain why this is a big deal https://t.co/Chr5ELTCsu https://t.co/nBJXq6lRKp" / X
- MatthewBerman on X: "MASSIVE Step Allowing AI Agents To Control Computers (MacOS, Windows, Linux) OS World allows agents to fully control computers, including MacOS, Windows, and Linux. OS World can benchmark agent performance like never before by giving agents a language to describe actions in a https://t.co/4fK82oDHiO" / X
- Questions for the Audience
- Quests for the Audience
- What is Langchain?
- LLM(Large Language Model) API(Application Programming Interface)'s
- Embeddings API
- Vector Databases
- LLM(Large Language Model) Prompt Chains
- RAG(Retrieval Augmented Generation)
- Chatbots
- Agents
- LLM Agents
- SQL | 🦜️🔗 LangChain
- Answer Questions using recursive queries
- Use recursion to troubleshoot queries
- Research Assistant
- Write an essay with one agent providing feedback and things to fix
- Troubleshoot code while computer programming
- SQL | 🦜️🔗 LangChain
- Review how to use OLLAMA's API
- Let's review Vector Databases
- Let's do a Vector Database Demo
- Weird Descriptions of Movies and or Anime
- Index a book, and find a passage describing a rich or poor character
- Sorting Characters in a book
- Order the characters in this book by
- Age
- Wealth
- Inteligence
- Importance
- Evil
- Good
- Religious
- Steps to calculate age
- Extract list of characters
- Search for the desired quality with the character, save each some where
- Order the characters in this book by
- Other Questions
- Next Steps
- Review RAG Tutorials
- Review Agent Documentation and Tutorials
- Build something cool
Intros
- What's your name
- What do you use AI for already
- What do you want to use AI for
- What do you want to build
My Answer
- Paul Mullins
- What do you use AI for already
- Replaced Stack overflow
- Use Recipes
- Summarize PDF documents
- What do you want to use AI for
- Quiz me on the media I consume
- Ingest data into a knowledge graph
- What do I want to build
- The Daemon
Accelerationisim
- We build shit
Outline
- What are LLMs (Large Language Model)
- What is a Vector
- What is an Embedding
- What is a Vector Database
- Text Chunking
- What is RAG
Langchain
- What is LangChain
- LLM API's
- Embeddings API
- Vector Databases
- LLM Prompt Chains
- RAG
Tutorial Steps
- Query Ollama or OpenAI
- Curl
- Langchain
- Document Loaders
- Text Files
- HTML Pages
- Vector Databases
- Vectora nodejs
bash
curl https://ollama.newatlantis.top/api/generate \
--location --insecure --request POST \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "mistral:7b",
"prompt": "How was the earth created?",
"stream": false
}'
curl https://ollama.newatlantis.top/api/generate \
--location --insecure --request POST \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "llama2-uncensored",
"prompt": "How was the earth created?",
"stream": false
}'