I used Claude 3.5 Project + Artifacts to help refactor most of the Ollama + Llama 3.1 #research project. I would call it a writing bot, but I'm not building it to automate writing. It's mostly a content wrapper around the Chat interface but it's good at generating code and building off of chat history.

work on codegen

work on codegen

writeup about the livecoding #codewithswiz

get up-to-date with codegen notes

played with symbex and llm to generate code. super cool and immediately useful

Experiment with fine-tuning GPT-3 for some programmatic SEO ideas #startupjobs

first steps with GPT-3 #life

mess with ChatGPT for generating code #japandev

try auto summarize the #nomads chat with ChatGPT with the new messages.db for potential 100% automated newsletter updates

#thecompaniesapi demo to my previous company again; work on codeblock that supports multiple languages/files and persists

added openai to create a seo description text to eacht topic #newsletterscan

🤖 playing around with GPT-3 #mylife

joined the GPT-4 beta and had it basically write a six chapter book for me.

experimenting with ChatGPT 4o

🤖 created a slackbot demo to try out a few ideas with #bots

updated some code that used OpenAI's now deprecated CODEX API, to convert some demos over to the ChatGPT api to see if things will still work.

🤖 I tried out the new ChatGPT Python API and gisted a demo gist.github.com/jefftriplett/…

submit codegen

🤖 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