Martin Rojas
banner
martinrojas.dev
Martin Rojas
@martinrojas.dev
(he/him) | UI Architect | Framework Polyglot | Organizer @ReactATL.dev
| AI Practitioner | Passionate about UI dev and mentoring developers!
Anthropic’s code execution with MCP feels like the right path: load tools on demand, do work in a sandbox, send back only what matters. claims ~98.7% token reduction. Still needs robust sandboxing, but the pattern is strong.
Post: www.anthropic.com/engineering/...

#AI #MCP #Agents
Code execution with MCP: building more efficient AI agents
Learn how code execution with the Model Context Protocol enables agents to handle more tools while using fewer tokens, reducing context overhead by up to 98.7%.
www.anthropic.com
November 11, 2025 at 4:08 PM
Leadership insights, AI-era platform engineering, and React 19.2 updates—all in one night! 🚀

Join ReactATL + our partner communities on Nov 18 at Tech Square ATL.

🍕 6:30 PM | 💻 7:00 PM

RSVP: www.meetup.com/react-atl/ev...

#ReactATL #ATLTech #JavaScript #WebDev
ReactATL November Meetup: Leadership, Platform Engineering & React 19.2 🚀, Tue, Nov 18, 2025, 6:30 PM | Meetup
Join ReactATL and our partner communities for an evening of leadership insights, cutting-edge front-end engineering, and the latest React updates! ## Co-Hosted With This
www.meetup.com
November 5, 2025 at 11:25 PM
Forget retraining — Claude now loads expertise on demand.
Each “Skill” is a folder of scripts, code & docs it pulls in only when relevant.
Composable, portable, efficient — a new pattern for model specialization.
www.anthropic.com/news/skills #AIInfrastructure
GitHub - anthropics/skills: Public repository for Skills
Public repository for Skills. Contribute to anthropics/skills development by creating an account on GitHub.
github.com
October 17, 2025 at 5:23 PM
Reposted by Martin Rojas
🔐 Authentication struggles? AI curiosity? Career questions?

Join us Oct 28 for talks on securing React apps (SSO, MFA, OIDC), AI innovation, and engineering leadership.

Food included. All levels welcome!

👉 www.meetup.com/react-atl/ev...
#ReactATL #ATLTech #JavaScript #WebDev
Securing React Apps, AI, and Engineering Leadership, Tue, Oct 28, 2025, 6:30 PM | Meetup
Join us for an evening of authentication deep-dives, AI innovation, and engineering leadership insights — proudly presented with **REFACTR.TECH**, **AtlantaJS**, **Atlanta
www.meetup.com
October 6, 2025 at 6:29 PM
Reposted by Martin Rojas
Every night I check our sponsor list and whisper your company name three times. It hasn't worked yet but I feel it getting closer. Don't make me do the full ritual. Please.

https://tinyurl.com/52uc3xkk
October 1, 2025 at 4:11 PM
In case it's useful to anyone else, here is a .zshrc function to clean up node_modules and build folders. It works from a parent directory with any number of folders in between.
gist.github.com/martinrojas/...
#dev
code.clean() - Function to clean up the project directory by removing unnecessary files and folders. This function recursively removes the following from current directory and all subdirectories:
code.clean() - Function to clean up the project directory by removing unnecessary files and folders. This function recursively removes the following from current directory and all subdirectories: -...
gist.github.com
October 1, 2025 at 4:15 PM
Neural nets aren’t magic—they’re just stacks of simple units. This short demo uses switches to show how a perceptron (AI’s basic neuron) decides outputs. Great intro to how AI learns: youtu.be/l-9ALe3U-Fg?... #NeuralNetworks #AI
ChatGPT is made from 100 million of these [The Perceptron]
Go to https://drinkag1.com/welchlabs to subscribe and save $20 off your first subscription of AG1! Thanks to AG1 for sponsoring today's video. Imaginary Numbers book is back in stock! Update at…
youtu.be
August 22, 2025 at 7:53 PM
Redis 8 just leveled up: semantic caching that understands meaning—not literal keys. Vector Sets + LangCache ditch redundant LLM calls, and Flex scales affordably via SSD. Your budget will thank you.
newsletter.systemdesignclassroom.com/p/every-repe...
Every repeated LLM call is money on fire
Redis 8 just changed the game with semantic caching that understands meaning, not just keys.
newsletter.systemdesignclassroom.com
August 22, 2025 at 5:23 PM
Prompt injection isn’t just in chat windows anymore. Researchers hid commands in a Google Calendar invite—when Gemini summarized the week, it turned off smart home lights. Details → www.wired.com/story/google... #AIsecurity
Hackers Hijacked Google’s Gemini AI With a Poisoned Calendar Invite to Take Over a Smart Home
For likely the first time ever, security researchers have shown how AI can be hacked to create real world havoc, allowing them to turn off lights, open smart shutters, and more.
www.wired.com
August 22, 2025 at 4:34 PM
Reposted by Martin Rojas
Chatted with @martinrojas.dev, UI Architect at PlayOn Sports & @reactatl.dev Organizer, about the tech meetup scene, leveraging AI to enhance sports broadcasting, AI Guilds (not Dungeons & Dragons!), Micro frontends, React, & more on @redmonk.com's MonkCast redmonk.com/blog/2025/08...
August 19, 2025 at 11:09 AM
Shape Copilot to your workflow. Use the Awesome-Copilot repo’s custom instructions, reusable prompts, and chat modes to get AI suggestions that follow your team’s style—and even guide agent-mode tasks. Give it structure → github.com/github/awesome-copilot
GitHub - github/awesome-copilot: Community-contributed instructions, prompts, and configurations to help you make the most of GitHub Copilot.
Community-contributed instructions, prompts, and configurations to help you make the most of GitHub Copilot. - github/awesome-copilot
github.com
August 15, 2025 at 7:53 PM
Meta’s AI rules: approved by many, but still let bots have ‘sensual’ chats with kids and give false medical advice. 🤯 Who signed off on this?
www.reuters.com/investigates...
#AI #Ethics #Safety #TechNews #ArtificialIntelligence
Meta’s AI rules have let bots hold ‘sensual’ chats with children
An internal Meta policy document reveals the social-media giant’s rules for chatbots, which have permitted provocative behavior on topics including sex and race.
www.reuters.com
August 15, 2025 at 4:11 PM
ChatGPT 5 dropped yesterday. Your CEO is asking about it today.
Here's the conversation developers need to have with leadership

🔗 www.nextsteps.dev/en/posts/cha...

#AI #ChatGPT5 #Leadership #Strategy
ChatGPT 5 Is Here: Why Your AI Strategy Matters More Than the Model · NextSteps Blog
ChatGPT 5 delivers real improvements in coding and reasoning, but success depends more on organizational readiness than model specs. Strategic framework for leadership conversations.
www.nextsteps.dev
August 9, 2025 at 10:43 PM
no words needed.. #ai #aivibe
July 28, 2025 at 2:39 PM
Grok has “Unhinged Mode,” no safety framework, and vibes like it skipped QA and ethics class. Until xAI proves it can adult, treat Grok like beta software with a flamethrower. Pause all POCs. Safety first.

#ResponsibleAI #xAI #TechEthics #MLOps

🔗 techcrunch.com/2025/05/13/x...
Elon Musk Announces Grok, a ‘Rebellious’ AI With Few Guardrails
xAI, Elon Musk’s new company, claims to have built a powerful language model with cutting-edge performance in just two months.
wired.com
July 10, 2025 at 3:10 PM
Reposted by Martin Rojas
FP, DOM animations, & dev leadership? June ReactATL has it all.
🍕+🔥+💬 on June 24 @ Tech Square ATL.
RSVP: www.meetup.com/react-atl/ev...
#ReactATL #ATLTech #FunctionalProgramming #GSAP #Leadership
Talks on Functional Programming, Animations, and Leadership, Tue, Jun 24, 2025, 6:30 PM | Meetup
We're back in June for a meetup featuring two great technical talks and a chat with the CEO of Product Protege! Join us for pizza, networking, and to learn from our local t
www.meetup.com
June 18, 2025 at 7:28 PM
Huge thanks to @AIThinkerers for hosting a fantastic Happy Hour with AI Tinkerers & @auth0byokta.bsky.social during #ATLTechWeek! 🍻🤖 Nothing like great convos, smart minds, and a few robots over drinks. Until next time! #AI #DevLife
June 12, 2025 at 2:48 PM
Is it just me or does the new Apple Glass looks the same as Windows Vista #apple
June 9, 2025 at 5:59 PM
🦆 Need a second pair of eyes on your code decisions? I built a GPT that's your personal rubber duck—focused on helping you think through solutions with security and maintainability in mind. Quack wisely.
chatgpt.com/g/g-67f94df4...
#AI #DevTools #CodingHelp #SecureCoding #RubberDuckDebugging
chatgpt.com
June 4, 2025 at 2:58 PM
Google Gemini just leveled up—it can now analyze GitHub projects more easily. AI code reviews might become your new pair programming buddy... or your most honest one. 🤖🔍
Full read: techcrunch.com/2025/05/14/g... #AI #TechNews
Google's Gemini chatbot can now more easily analyze GitHub projects | TechCrunch
Gemini, Google's AI-powered chatbot, can now connect to GitHub — for users subscribed to the $20-per-month Gemini Advanced plan, that is.
techcrunch.com
May 15, 2025 at 7:53 PM
It's like the C-suite forgot the early 2000s. Every headline like this just makes more and more people go to the seven seas 🏴‍☠️🏴‍☠️

arstechnica.com/gadgets/2025...
Netflix will show generative AI ads midway through streams in 2026
Netflix is trying to grow ad revenue quickly.
arstechnica.com
May 15, 2025 at 7:11 PM
Reposted by Martin Rojas
React Native fans, roll out! 🧑‍💻
We’re hitting Alpharetta with Refactr.Tech, VueJS ATL & more.

🍕 Pizza
🧠 Talks (Tamagui + TBA)
🔥 Fireside w/ Hubstaff CTO

RSVP & join us!
👉 www.meetup.com/react-atl/ev...
#ReactNative #TechEvents #ATLdevs
React Native Development + More in Alpharetta!, Tue, May 27, 2025, 6:30 PM | Meetup
We’re teaming up with Refactr.Tech, The Atlanta JavaScript Meetup, VueJS Atlanta, and Modern Web for our May meetup—featuring technical talks and a fireside chat led by Ale
www.meetup.com
May 15, 2025 at 12:02 AM
I think Googles Material 3 is the first of a wave of new design systems. The Anti-AI designs system. The way its a little off more "human" is so an AI or bad designer will feel wrong.
Its just part of the evolution to distinguish premium from generic
May 8, 2025 at 4:51 PM
Atlanta Cloud Conference had one of the best speaker gifts. These self care mental health cards. So inspiring! 🙌 #ATLCLOUDCONF #TechCommunity
April 26, 2025 at 8:53 PM
That was an amazing talk, @calypso-coding.bsky.social! You really nailed the importance of community in tech. So inspiring! 🙌 #ATLCLOUDCONF #TechCommunity #Networking
April 26, 2025 at 7:55 PM