Super valuable! Could even export your best chats, feed it into code interpreter, and ask for the prompts that provided best responses/your favorite (most used) prompts Paste it into a notion database, and share across teams to optimize everyone’s workflows!
AUTOMATION
-
OpenAI Launches Memory Feature for ChatGPT Users
By
–
OpenAI just launched a new memory feature to select users, enabling ChatGPT to remember key details from prior chats. Users can control what's remembered, with options to delete specifics, wipe memory, or disable the feature.
-
Weekly cleaning burden sparks discussion on home automation
By
–
Who has time to clean weekly? Are you people octopuses?
-
Can AI Assistants Handle DMV Administrative Tasks?
By
–
when can I send https://
huggingface.co/chat/assistants to the DMV for me? -
AI Driving More Layoffs Than Companies Want to Admit
By
–
#AI Is Driving More Layoffs Than Companies Want to Admit https://
finance.yahoo.com/news/ai-drivin
g-more-layoffs-companies-174840542.html
… -
Outlook and Superhuman Develop Advanced Email AI Features
By
–
Supposedly Outlook is working on this and @Superhuman can already kind of do this.
-
AI Laundry Folding Robots Coming in a Few Years
By
–
There are robots that'll do this. Give it a couple years and you'll have an AI laundry folding robot. 🙂
-
Beyond Document RAG: Context Routing for Better Chatbot UX
By
–
This is cool but also shows how document RAG alone is not enough.
— Aaron Ng (@localghost) 13 février 2024
Dragging in meeting notes to get a restaurant from a transcript is weird and clunky. Some sort of local data-store or context routing layer is still a key missing part of chatbot ux. https://t.co/mj6qu1HGVJThis is cool but also shows how document RAG alone is not enough. Dragging in meeting notes to get a restaurant from a transcript is weird and clunky. Some sort of local data-store or context routing layer is still a key missing part of chatbot ux.
-
LangGraph Enhances Plan-Execute Agents with Flexible Code Control
By
–
6/ This all extends our previous work on plan-and-execute agents. But now with LangGraph, it's more flexible than ever – you own the code!
-
Plan-and-Execute Agents: Faster LLM-Powered Execution
By
–
Plan-and-Execute Agents We've added new documentation on how to make plan-and-execute style agents using LangGraph. Relative to older agent designs, these agents promise: Faster Execution: fewer calls to large models, and execution of tools while the LLM is still