Back
Similar todos
🤖 worked on a better template/pattern library script tonight 
#mylife
🤖 dove a little deeper into a screenshots to code tool and the results are getting much better 
#research
🤖 played around with adding extra context in some local Ollama models. Trying to test out some real-world tasks I'm tired of doing. #research
🤖 played with Aider and it mostly working with Ollama + Llama 3.1 #research
📝 prototyped an llm-ollama plugin tonight. models list and it talks to the right places. prompts need more work.
some fun coding with cursor, i actually ended up benefitting from cursor-small too 
#life
🤖 built a new code generator tool when I wasn't driving today. #mylife
🤖 more working with Ollama and Llama 3.1 and working on a story writer as a good enough demo. #research
fix issues for most recent tester
played with symbex and llm to generate code. super cool and immediately useful
🤖 added more meta options to my code writing robot. #mylife
🤖 more robot work, but figured out how to get better results limiting num_predict for git commit messages #research
try combine Segformer object/wall detection with ControlNet depth mask so constructions stays intact but interior can get changed 
#interiorai
🤖 spent some time getting Ollama and langchain to work together. I hooked up tooling/function calling and noticed that I was only getting a match on the first function call. Kind of neat but kid of a pain.
🤖 Created an Ollama + Llama 3.2 version of my job parser to compare to ChatGPT. It's not bad at all, but not as good as GPT4. 
#jobs
🤖 I realized I didn't need to chain multiple tool calls together for my Calendar Agent to work. I can have my create check and then stop overthinking it. #research
🤖 testing Claude 3 for some low hanging fruit code writing tasks
🤖 played with Ollama's tool calling with Llama 3.2 to create a calendar management agent demo #research
🤖 wrapped my brain around proxies and built an app which collects and tests them #mylife