Founders

Co-founder at Agora. Building governance software. Talk less, smile more.

I am routing based on task or even sub task!

Genuinely — never thought about the models with a split model. Do you find yourself choosing one depending on the task you are working on? For me its ChatGPT for all.

thot leadership at /opacity --
host of vibe check podcast 
▶️ https://pods.media/vibe-check
❄️ icebreaker.xyz/dawufi

Notice that Siri is not on this list 🍎

gemini- trying to replace google assistant but can't do the one thing i currently use assistant for (setting reminders)

someone at google needs to be fired or this

I usually stick to Claude and Gpt. 

Claude is very good at writing code -given it knows the documentation.

GPT is good for research purposes, general discussion and writing. I love the ‘search’ toggle on it and can’t emphasise how much time it saves me.

pretty much the same 
i actually have started using agent routing workflows a bit - let an LLM decide which agent to call

Agree with all of these. My add-ons:

- Gemini is great for large context and access to YouTube and Search

- Grok has access to tweets which is very useful

- OpenAi released a bunch of new stuff that changes a lot potentially (4.1 has 1M context now, o4, o3-mini)