Skip to main content

BrazilJS : And a wall of dreams



Today I want to talk a little bit about BrazilJS. The memories are still fresh since it happened just a month back for an action packed 2 days at the beautiful city of Porto Alegre. 

 Apart from having some of the most awesome speakers I have heard in a conference it had one of the most active and engaged audiences I have seen.

Mozilla had a remarkable presence there, including the captivating talk of Mike Taylor about "How to make browsers compatible with the Web” which I highly recommend you to listen to if you are a web developer or aspire to be one. You will not only benefit yourself from it, but will do a favour to everyone who tries to follow web standards. He gave the same talk in Viewsource just a few weeks after BrazilJS.

 Apart from the talk we had a WebVR and a HTML Game booth set up at the venue, which kind of looked like this when we were setting it up
The WebVR booth at BrazilJS
The item up for demo was “Puzzle Rain”. A WebVR game specially tuned for HTC Vive and utilizes it’s controller. In short what you do in the game is to collect different shaped pulsating cubes (each associated with a different tune, so this really is an audio-visual experience) and put them in their appropriately shaped hovering boxes. Each time you perform it properly the tune changes, at the end giving you a soothing feeling of a musical journey.
What was more awesome than the demo and long line in front of the demo booth was the reaction of people while they played the game. Which made them look something like this
Play in style!
And remember the long line I talked about?

And when you think everything is over we will always have an awesome performance



That was Nolan Lawson, and how can a web conference complete without Microsoft (Edge) and Mozilla going side by side

And when you say that, how can Google be left behind?
Presenting Mozilla, Microsoft and Google pledging to make the web a better place.
Mozilla, Microsoft and Google

But that still is half of what I wanted to write. By now you may have noticed from the pictures that we had an awesome glass wall at the booth. The Brazillian Mozilla community came up with the awesome idea to invite everyone, passionate bout open web to write what they want from the open web and Mozilla on the wall. And the audience, participants, the community as a whole took it up as we see here


Which transformed the wall into something like this

I like to call this a “Wall of dreams”. There are wishes, dreams written in this wall that represents the open source citizens. The “Dumbledore’s army” that roots for open web, a web outside the walled garden, roots for Mozilla to win.
There are wishes written in 7 languages in that wall (including my own which I wrote in Bengali). Can you spot all of them? :)

 That’s a task I will leave to the readers.
And at the end this is crew which kept it going in the booth


And how can we miss your hero Gabriel!!!
Oh and with the whole team who made BrazilJS and every part of the booth possible
The Mozilla Brazil Community
Of course after all these you need a super strong shot of Expresso Rust
And that’s how it all went. At the end, of-course our very own superstar Michael Ellis had to steal the show

This is how we wrapped up #BrazilJS


Comments

Post a Comment

Popular posts from this blog

Racecraft (Project Koru) · Prologue — The Origin Story

Racecraft · Prologue , The Origin Story It Started With a Wine List and a Question About Racing How a happy-hour conversation in the Bay Area turned into a trustable AI race coach , and then into a second version that runs entirely on a phone, on the NPU. This is the prologue to a five-part series. Two years ago(1st November, 2024) I was in the Bay Area for a GDE Summit. If you've never been: it's a couple of days of talks among Google Developer Experts, the kind of people who get unreasonably excited about a new on-device runtime, and then , mercifully , a happy hour where everyone stops performing and just eats. We ended up at a restaurant(Puesto Santa Clara), a long table of GDEs, and I was doing the most important engineering of the evening: trying to decide which wine to order. Across the table was Ajeet Mirwani . I don't even remember how the wine talk turned into racing talk , these things drift , but the moment the word "racing" ...
Post a Comment
Read more

The Throughput Trap: Benchmarking vLLM on OpenXLA and the Reality of Production LLM Serving

vLLM Systems · DevLab 2026, Deep Dive I was recently invited by the Google TPU team to speak at the OpenXLA Summer DevLab 2026 . This post breaks down our deep-dive evaluation of the matured vLLM + OpenXLA stack, the fundamental engineering mismatches between CUDA and XLA serving paths, and why traditional capacity metrics are lying to you. If you are operating large language models at enterprise scale right now, your platform architecture team is likely staring at a massive infrastructure crossroads: Should we migrate our core serving workloads from GPUs to TPUs? Historically, NVIDIA's CUDA ecosystem was the only serious option for user-facing, low-latency LLM generation. But here in 2026, the economics and infrastructure options have transformed. Google TPUs are highly available, cheaper per chip, and the open-source serving stack built around vLLM and OpenXLA has officially achieved absolute production readiness. Yet, when our infrast...
Post a Comment
Read more

A Split‑Brain Neuro‑Symbolic Training Method for High‑Velocity Autonomous Coaching from Telemetry

 Author: Rabimba Karanjai Scope: Problem statement + data methodology + model training (no deployment discussion) Abstract Real‑time coaching in motorsport is a safety‑critical learning problem : a system must map noisy, high‑frequency telemetry to short, actionable guidance that remains physically consistent and avoids hazardous recommendations . This paper proposes a “Split‑Brain” training formulation that separates (i) a semantic coaching target (what action/critique should be expressed) from (ii) a reflexive interface (how actions are represented as compact, verifiable tokens). The approach trains a Small Language Model (SLM) in the Gemma family [1] using QLoRA fine‑tuning [2] , and introduces a telemetry tokenizer plus teacher‑student synthesis pipeline to generate instruction‑action pairs at scale. Core contribution: a reproducible method to convert “ golden lap ” differential tel...
Post a Comment
Read more