Sign in

Sergio Garcia Murillo

@murillo2000.bsky.social
324 followers 380 following 708 posts

Tech enthusiast and WebRTC expert, father of Leo and Mia. Mathematician

PostsRepliesMedia
Sergio Garcia Murillo @murillo2000.bsky.social · 24/09/2026
wondering if my free quota resets have anything to do with my naive background tasks trying to resolve the Riemann Hypothesis? 🤣
000
Sergio Garcia Murillo @murillo2000.bsky.social · 15/09/2026
A los trolles si no les haces casito se mueren.. 🤣
061
Sergio Garcia Murillo @murillo2000.bsky.social · 03/09/2026
ChatGPT is down!
120
Sergio Garcia Murillo @murillo2000.bsky.social · 12/08/2026
251
Sergio Garcia Murillo @murillo2000.bsky.social · 12/08/2026
000
Sergio Garcia Murillo @murillo2000.bsky.social · 12/08/2026
000
Sergio Garcia Murillo @murillo2000.bsky.social · 12/08/2026
Ready!
010
Sergio Garcia Murillo @murillo2000.bsky.social · 30/07/2026
This is an outrage! (If you know, you know)
000
Sergio Garcia Murillo @murillo2000.bsky.social · 30/07/2026
ChatGTP destkop app has a customizable overly pet.. now we are talking 😂
000
Sergio Garcia Murillo @murillo2000.bsky.social · 29/07/2026
"No se puede desbrozar en el monte porque hay muchas trabas administrativas" por que sera Jose Luis..
001
Sergio Garcia Murillo @murillo2000.bsky.social · 10/07/2026
😱😱😱😱😱 m.youtube.com/watch?v=j5z5...
m.youtube.com
Miniature Masters: Worlds First miniature & Warhammer painting TV-Show EP1
YouTube video by Squidmar Miniatures
000
Sergio Garcia Murillo @murillo2000.bsky.social · 17/06/2026
It feels that I am again in a mood of cleaning inbox, unsubscribing mail list, groups, slacks channels.. Feels sooo good
static.klipy.com
Marie Kondo Tidying Up Clothes
ALT: Marie Kondo Tidying Up Clothes
020
Sergio Garcia Murillo @murillo2000.bsky.social · 18/04/2026
90s scape room
010
Sergio Garcia Murillo @murillo2000.bsky.social · 11/04/2026
Is agile still valid with AI development? A lot of “Agile” today is really just project management wrapped around implementation details. That made sense when writing code was the expensive part and a bottleneck. With AI, implementation gets cheap fast. Review, judgment, and validation do not.
100
Reposted by Sergio Garcia Murillo
Jess Rose @jesslynnrose.bsky.social · 04/03/2026
Global remote, 🚨paid🚨 open source grant program from @igalia.com for folks to learn more about Linux, Graphics, JS DevTools, Multimedia + GStreamer or Web Standards. - apply by April 3 - €7k for 450 hours over 3-6 months - uni students or self directed learners www.igalia.com/coding-exper...
igalia.com
Igalia Coding Experience | Igalia - Open Source Consultancy and Development
Igalia is an open source consulting firm specialised in the development of innovative projects and solutions. Our engineers have expertise in a wide range of technological areas, including browsers an...
11729
Sergio Garcia Murillo @murillo2000.bsky.social · 07/02/2026
“Refactoring as escape” is an anti-pattern that appears when a new team takes over a running service. Instead of owning its current behavior, failures are blamed on the code, and large refactors are proposed in place of understanding root causes.
221
Sergio Garcia Murillo @murillo2000.bsky.social · 04/12/2025
I’ve been playing with LLM-generated code and didn’t feel comfortable releasing it under OSS license without disclaimers and traceability So… introducing the VIBE License, a permissive option for LLM generated code, with straight safeguards and provenance requirements github.com/murillo128/v...
github.com
GitHub - murillo128/vibe-license: VIBE Permissive License for LLM generated code
VIBE Permissive License for LLM generated code. Contribute to murillo128/vibe-license development by creating an account on GitHub.
010
Reposted by Sergio Garcia Murillo
Gianni Rosato @giannirosato.com · 22/11/2025
Enhancing HDR on Instagram for iOS With Dolby Vision: engineering.fb.com/2025/11/17/ios/e… #av1 #vp9 #hdr
021
Sergio Garcia Murillo @murillo2000.bsky.social · 18/11/2025
Yesterday I asked ChatGPT to refactor some code and it said: Look, this is way too complex and I’m just going to introduce bugs, so I’m not doing it. But here’s how you could do it yourself instead. The way it slipped out of the task while sounding confident and helpful.. We did it, we built AGI
141
Reposted by Sergio Garcia Murillo
Sergio Garcia Murillo @murillo2000.bsky.social · 08/11/2025
I think I am going to use this meme quite a few times in the future 🤣
143
Sergio Garcia Murillo @murillo2000.bsky.social · 08/11/2025
I just attended my firsts MoQ meetings during ETF 124, here are my personal thoughts and recap: www.linkedin.com/posts/sergio...
linkedin.com
I just attended my firsts MoQ meetings during ETF 124, here are my personal thoughts and recap: My overall impression is that the MoQ Transport draft may be less stable than I expected, with several…...
I just attended my firsts MoQ meetings during ETF 124, here are my personal thoughts and recap: My overall impression is that the MoQ Transport draft may be less stable than I expected, with several ...
150
Reposted by Sergio Garcia Murillo
webrtcHacks @webrtchacks.com · 04/11/2025
The #MoQ vs. #WebRTC debate has been missing use case details. See our new post for an analysis of: * 1:1 calls * #VoiceAI * Meetings * Streaming * Live streaming * Webinars / Townhalls
webrtchacks.com
WebRTC vs. MoQ by Use Case - webrtcHacks
WebRTC vs. MoQ by use case—1:1 calls, Voice AI, meetings, and live 1→many. See where MoQ, WebRTC, and HLS each win with details.
043
Reposted by Sergio Garcia Murillo
webrtcHacks @webrtchacks.com · 07/10/2025
New post: Is everyone switching to #MoQ? Fippo responds to some FUD about #WebRTC in a recent MoQ launch post with some historical perspective & Chrome Status data
webrtchacks.com
Is everyone switching to MoQ from WebRTC? - webrtcHacks
Media over QUIC (MoQ) hype check: WebTransport usage data with a rebuttal to Cloudflare’s post—why no one has switched to MoQ from WebRTC.
155
Sergio Garcia Murillo @murillo2000.bsky.social · 21/08/2025
The real problem is not people asking AI for random (potentially dangerous) stuff, but when it contaminates so many sources that an educated person can't differentiate the legitimate from the hallucinated versions
040
Sergio Garcia Murillo @murillo2000.bsky.social · 21/08/2025
This
010
Sergio Garcia Murillo @murillo2000.bsky.social · 30/06/2025
Ueno - Tpkyo
010
Sergio Garcia Murillo @murillo2000.bsky.social · 30/06/2025
Ueno - Tokyo
000
Sergio Garcia Murillo @murillo2000.bsky.social · 28/06/2025
Team Lab Borderless - Tokyo
010
Sergio Garcia Murillo @murillo2000.bsky.social · 20/06/2025
This feels sooo goood 🥳
140
Sergio Garcia Murillo @murillo2000.bsky.social · 07/06/2025
Los más grandes
000
Sergio Garcia Murillo @murillo2000.bsky.social · 07/06/2025
Alcala Norte!
000
Sergio Garcia Murillo @murillo2000.bsky.social · 20/05/2025
🤦‍♂️
010
Sergio Garcia Murillo @murillo2000.bsky.social · 15/05/2025
Our new foster kittens are just lovely
000
Sergio Garcia Murillo @murillo2000.bsky.social · 08/05/2025
alias dicker="docker" 🙈 I am the only one making this typo constantly?
110
Reposted by Sergio Garcia Murillo
Hugh Janus @eyeinthesky8.bsky.social · 10/04/2025
The ketamine kid.
0405
Reposted by Sergio Garcia Murillo
Dave @allcaps.bsky.social · 03/04/2025
Interesting theory. The tariffs feel very AI generated and using a source list of ISO 3166 country codes tracks. It also demonstrates the lack of any thought that went into this scheme (which is currently tanking US and international markets).
132
Sergio Garcia Murillo @murillo2000.bsky.social · 30/03/2025
For the english audience, Google is providing as top result a completely bogus answer AI generated answer about how to write a number in spanish. One of the "sources" is a web page translated from english about english numbers!
120
Sergio Garcia Murillo @murillo2000.bsky.social · 30/03/2025
Pero que mierda es esta
010
Sergio Garcia Murillo @murillo2000.bsky.social · 27/03/2025
WHIP has just been published as RFC9725! www.rfc-editor.org/rfc/rfc9725....
rfc-editor.org
RFC 9725: WebRTC-HTTP Ingestion Protocol (WHIP)
This document describes a simple HTTP-based protocol that will allow WebRTC-based ingestion of content into streaming services and/or Content Delivery Networks (CDNs). This documen...
070
Sergio Garcia Murillo @murillo2000.bsky.social · 23/03/2025
just wow 🤯
media.tenor.com
a black cat laying on the floor with its eyes glowing
ALT: a black cat laying on the floor with its eyes glowing
000
Sergio Garcia Murillo @murillo2000.bsky.social · 05/03/2025
I killed a guy in fortnite with a tesla cybertruck today. Obviously I destroyed the Cybertruck afterwards and another player come to dance with me while it exploded 🤣
120
Sergio Garcia Murillo @murillo2000.bsky.social · 28/02/2025
me ha encantado esta reseña, ni que la hubiera escrito @estigabi.bsky.social 🤣
000
Sergio Garcia Murillo @murillo2000.bsky.social · 21/02/2025
proof that you can be rigth randomly with incorrect arguments
131
Reposted by Sergio Garcia Murillo
Patricia Aas 🐢🏳️‍🌈 @patigallardo.bsky.social · 29/01/2025
Finally getting around to watching Helge Pennes talk on C++ and security 😎 (someone tell him to get on here or tell me his @) youtu.be/gG4BJ23BFBE?...
youtu.be
The existential threat against C++ and where to go from here - Helge Penne - NDC TechTown 2024
YouTube video by NDC Conferences
2253
Sergio Garcia Murillo @murillo2000.bsky.social · 28/01/2025
ok, here we go...
000
Sergio Garcia Murillo @murillo2000.bsky.social · 27/01/2025
para que quieres saber eso? jajaja saludos
110
Sergio Garcia Murillo @murillo2000.bsky.social · 27/01/2025
Some interesting insights in DeepSeek
010
Reposted by Sergio Garcia Murillo
Json Momoa @jotatx.eurosky.social · 27/01/2025
the real issue is that to achieve this llm model the researchers only needed 2% of the average cost to train new models of similar number of tokens that starts at 1.5 billion tokens and goes up to 671billion tokens. released under MIT license, btw api-docs.deepseek.com/news/news250...
License
The model weights are licensed under the MIT License. DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that:

The Qwen distilled models are derived from Qwen-2.5 series, which are originally licensed under Apache 2.0 License, and now finetuned with 800k samples curated with DeepSeek-R1.

The Llama 8B distilled model is derived from Llama3.1-8B-Base and is originally licensed under llama3.1 license.

The Llama 70B distilled model is derived from Llama3.3-70B-Instruct and is originally licensed under llama3.3 license.
2214
Reposted by Sergio Garcia Murillo
Led By Donkeys @ledbydonkeys.org · 22/01/2025
Tesla Gigafactory, Berlin (Collaboration with @politicalbeauty.bsky.social )
Picture of the Tesla gigafactory in Berlin with Elon Musk projected and the words Heil Tesla
279113323868
Sergio Garcia Murillo @murillo2000.bsky.social · 10/01/2025
20 años después..
020