Grad school applications are stressful! I shared *practical* tips from my experience at Microsoft Research last year and my slides were greatly appreciated. To encourage inclusivity in academia, I am making my slides public: https://t.co/NuA8er5nB6 #AcademicChatter @ThePhDPlace

AI commentary unlocks new possibilities - Personalized commentary: beginner-friendly, numbers-focused, drama-heavy, biased to your team...your choice! - Interactive: e.g. discuss strategy with the commentator as the game unfolds - Democratizing sports: commentary can drive

I've been thinking about AI applications beyond chatbots and coding agents. Here's one we came up with: AI-generated sports commentary! Check it out:

My hot take on AI browsers: if it's Chromium-based (@perplexity_ai @OpenAI), I'll try it and move back to Chrome. Too much cognitive investment to leave behind (and Chrome will ship the same features anyway). The only way to dethrone Chrome is to rethink browsing – from scratch!

If you code in python and aren't using uv yet, this is your sign to jump ship! It's so fast and fun...I look forward to installing dependencies now lol @astral_sh

This @acm_chi paper finds that AI assistance homogenized writing in a controlled experiment with American and Indian participants (n=118). With a classifier, it was harder to predict writer origin without AI (83.5% vs. 90.6%). https://t.co/4U3yycxE2E

("vibe-coded" this with @cursor_ai over the weekend haha; thanks to @karpathy for the motivation!)

9. 🎉Minimize boilerplate setup. Maximize productivity and rapidly build stuff! 📦 pip install miniai 🔗 https://t.co/YnfSFTZ13r Feel free to request more features and contribute!

8. ✨ Function decorator: The magic wand Turn ANY function into an AI-powered function. You can now write natural Python-looking code while incorporating AI capabilities.

7. 🔍 Customizable Usually you’ll just get the processed response (what you probably need). But if you need the raw response for more involved use cases, that's also a param away.

6. 🤖 Helper functions for common tasks We shouldn't need an elaborate setup for such simple tasks.

5. 🛠️ Specific Use Cases: Already handled AI providers have completely different APIs for embeddings, text-to-speech, speech-to-text, etc. They shouldn't be so hard to reach.

4.⚙️ Hyperparameters Anthropic and OpenAI handle something as basic as system prompts differently. MiniAI creates a unified interface. If you want it, say it and it'll all work.

3. 🔄 Flexible: The defaults will usually work. But need a different provider? A different model? A param away.

2. 🎯 Intuitive: It's designed to...just work

1. It's hard to prototype when you have so much setup to do. MiniAI is designed to fix that:

🚀 Introducing MiniAI - AI made ridiculously simple! Sick of OpenAI boilerplate? Tired of writing complex chains to do simple AI tasks? MiniAI is a minimal, flexible solution for byte-sized AI tasks. pip install miniai Examples ⬇️

PhD students: Upon paper rejection: "I got very harsh reviewers. I don't deserve this." Upon paper acceptance: "I probably got lenient reviewers. Do I deserve this?" #ImposterSyndrome

I miss my "young hacker" era, when I could think of an idea and implement it within a few hours — without being bogged down by a lit review, competitor analysis, the "right" approach, etc. Just building for the joy of building! Did that this holiday szn and felt refreshed :')

Just attended my first (big) conference and couldn't get over the feeling of people saying, "I've read your papers!" Felt giddy in my stomach every single time! Small wins of academia ✨

Excited to head to CSCW next week with @FarhanaShahid_, @SharonHeung, @agdhruv, and Joy Ming where we will be presenting work on Human-AI Interaction and Social Computing! If you're attending CSCW and would like to connect, please reach out.