Still buzzing from @europython 2026! 🐍 I presented “Python Games in the Browser,” met Guido van Rossum, experienced Core.py IRL, connected with the DSF community, and joined the sprints and socials. An unforgettable week with the Python community! 💛💙

It was so much fun to discuss compilers, python and wasm. Major kudos to @kranirudha for doing such amazing work on waspy. I think we can definitely make shipping wasm with python easy.

This was really interesting to present and let people know that waspy exists 😎

I am looking forward to this XD

Had a great time talking about building 3D worlds… So great to find folks working on incredible stuff in the area… at @IndiaFOSS 2026 ⚡️

Sonnet 5.5 better than Opus 5.5 on agentic coding 😮

TIL: Been using Linux’s fg command a lot lately. Ctrl+Z suspends the current process. jobs lists suspended jobs. fg brings it back. Great when using agents/editors: Ctrl+Z → run a quick command → fg No extra pane or tab. Tiny command, great workflow. Follow for more such TIL

Runek is an amazing project I think this talk is going to be awesome. Looking forward to it 😎

And here we go ....

Every poor soul or compute needs some motivation XD This somehow reminds me of Marvin

Are we really turning into a meat-proxy? I am a strong advocate to use AI agents for the job but to know what exactly is going on, I would rather think that we are getting better at describing what we want and how we want it. We are getting better at refining the requirements.

The recording of my #EuroPython talk is now available! 🎥🐍 If you missed it at the conference, you can catch the full session here: ▶️ youtu.be/7vKx1hx9m9s Would love to hear your thoughts and feedback! cc: @europython @europythons #Python #OpenSource

Prepped another board with the CH32V203 and edited a process video !! Pasting, assembly, reflow, .. and USB comms testing

We're excited to announce our next speaker for import_bengaluru - Ganesh Kathiresan who will be representing @PyData BLR and be speaking on "How NumPy divides integers really fast (and the bugs I caused doing it)" Ganesh is a SSE at Amazon and an OSS contributor to @numpy_team

A single GPU serving an LLM does not process one request at a time. It processes one step at a time, and a step can belong to any request that happens to be ready. Here's how it goes ... Generating a response is not one big computation; it is a loop. Each iteration of that loop (a step) takes the current sequence, runs one forward pass, and produces exactly one new token. So, a response of 500 tokens is 500 separate steps. This is where continuous batching kicks in. Continuous batching decouples the step from the request. After every single token generation step, the scheduler checks the batch. Any sequence that has finished (hit an EOS token) is evicted immediately, and any new request waiting in the queue is slotted into that now-open slot, right in the middle of everyone else's generation. So at any given moment, the GPU's batch might contain token 3 of a brand new request sitting right next to token 400 of a long-running one. The GPU does not know or care; it just runs a forward pass over whatever sequences are currently assigned to its batch slots. Now, this is only possible because attention and the KV cache are computed per-sequence. Each request's keys and values live in their own memory region (this is where PagedAttention comes in, to manage that memory efficiently), so batching different requests together at different stages of generation does not corrupt anyone's context. Thus, GPU utilization stays high and average latency drops, because no one is stuck waiting behind a single slow request that happened to start first. Hope this helps.

This is more for agent builders... Primarily coding agents but extends to more general agent builders too. I imagine some amount discomfort in not knowing where your agent is running, which some agents use with docker and other directly on your system. Was working on something similar and ended up writing about it here: blog.anirudha.dev/give-your-…

Excited to announce @InMobi as our Venue Partner for #import_bengaluru! 🎉 Thanks to the InMobi team for supporting Bengaluru’s Python & open-source community. 📅 6 Sep | 10 AM–6 PM 📍 InMobi, Bengaluru RSVP: luma.com/5v2wfu9f