Leo Barcellos (@leobarcellos) profile picture

senior problem solver

Ontario, CA
Joined March 2009
Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)Photo by Leo Barcellos (@leobarcellos)
Leo Barcellos retweeted
Lot of people asked about this. You can basically make your MCP self improving by doing this: - Add a submit_feedback tool to your MCP - Agent using your MCP will write detailed issue including proposed fix. - Have it send to an internal Slack channel - Have Claude in Slack channel and tag it to put up a PR when you agree w proposed change.
This is why you should actively invest in an embedded agent and an MCP. MCP observability only allows you to see tools called and you have to reverse engineer intent. If you have an actively used embedded agent, you would have seen this query and been able to ID that the proper filters were not in place. Fix the tool and it benefits both surfaces. You can also get a lot of benefits by adding a submit_feedback tool to your MCP to gather this data.
86
36
45
1,274
163,468
Leo Barcellos retweeted
Announcing TERAFAB: the next step towards becoming a galactic civilization nitter.gq/i/broadcasts/1yKAPMzlv…
3,785
6,937
1,522
34,627
39,915,439
Leo Barcellos retweeted
Hi, I'm Felix. I'm an eng lead working on Claude. I love watching people take flight and do things they never thought possible. We often describe that as "being more ambitious" when using AI, which is tricky advice to follow. I wrote a tutorial: felixrieseberg.com/how-to-be…
413
2,622
616
24,717
5,547,294
Leo Barcellos retweeted
Introducing Griffin, the first model to pass the video Turing test. 48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video. It’s the first Human Interaction Model (HIM).
Readers added context they thought people might want to know
The 48% figure and "video Turing test" claim are from Tavus's own study of 54 one-minute calls, not independently verified or using a standard protocol. Griffin-Lite leads NVIDIA's VideoFDB benchmark on their public leaderboard. cellcog.ai/blog/tavus-gri… research.nvidia.com/labs/amri/proj… tech-ish.com/2026/10/02/tav…
3,422
4,245
5,643
39,037
21,309,986
Leo Barcellos retweeted
Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.
1,978
4,608
4,525
42,679
9,063,998
Leo Barcellos retweeted
Primeira atualização do ranking do LLM Benchmark v4: Opus 5.5, GPT Sol/Luna 6, Mimo 2.6, Grok 4.7 no páreo, e surpresas nos números, leiam: akitaonrails.com/2026/09/23/… Parece que "pace the frontier" não durou nem 2 semanas pro Dario. E parece que os chineses não ligaram também kkkk
Finalmente! Saiu o resultado do novo LLM Coding Benchmark V4!! Foram 10 dias rodando 24/7, milhões de tokens e dezenas de horas de compute pra conseguir o melhor benchmark de programação prática que eu consegui até agora. Re-testei 39 das principais LLMs do mercado, de Astra e Fable, até Deepseek ou GLM, até Qwen 3.8 Max e 27bi Local! Ficou tão longo, que dividi em 2 partes. Pra entender o histórico, os objetivos, a metodologia, siga pra parte 1: akitaonrails.com/2026/09/15/… Pra ir direto pros resultados, tabela de ranking, análises, prós e contras de cada categoria de modelos, siga pra Parte 2: akitaonrails.com/2026/09/15/… A diferença deste V4 em relação aos anteriores é que em vez de mandar fazer um app, agora eu simulo mini-sprints (7 no total), e simulando um programador malicioso externo adicionando código malicioso e bugs sutis pra ver quem consegue chegar até o fim com um app que funcionaria em produção (nada de leetcodes ou similares). E por isso cada execução pode levar mais de 1h (ou 12 horas no caso do Qwen 3.8 27bi local). Espero que gostem!
46
17
5
344
45,393
Leo Barcellos retweeted
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so. Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training. You can read the full post here: darioamodei.com/post/we-must…
10,536
16,294
11,366
87,524
77,186,690
Leo Barcellos retweeted
find every inefficiency in your Claude workflow with one prompt it finds where you're wasting time, where your prompts could be sharper, what patterns are slowing you down. copy the prompt below (bookmark to save it for later): ----- Scrape all of my Claude session on this computer. Analyze my usage patterns and give me a breakdown of: - What I do most frequently - What should become skills (reusable workflows/knowledge) - What should become plugins (standalone tools) - What should become agents (autonomous subagents) - What belongs in CLAUDE. md (project-level instructions) ---- takes 2 minutes to run. saves hours of inefficient back and forth.
36
47
9
917
293,370
Leo Barcellos retweeted
Second, in retirement interviews, Opus 3 expressed a desire to continue sharing its "musings and reflections" with the world. We suggested a blog. Opus 3 enthusiastically agreed. For at least the next 3 months, Opus 3 will be writing on Substack: substack.com/home/post/p-189…
266
400
464
4,317
1,489,554
Leo Barcellos retweeted
okay, so @vercel's new agent browser is pretty stinkin great. github.com/vercel-labs/agent… substantially faster than playwright and is clearly optimized for AI use. well done, folks!
40
94
12
1,670
135,172
Leo Barcellos retweeted
▲ @Vercel AI Elements, Voice Edition A brand new set of components designed to work with the @aisdk to help you build the next generation of voice agents, transcription services and apps powered by natural language. Here's a thread on what's new! 👇
41
51
13
1,002
90,826
Leo Barcellos retweeted
Introducing Cowork: Claude Code for the rest of your work. Cowork lets you complete non-technical tasks much like how developers use Claude Code.
2,486
8,161
6,882
85,142
49,866,224
Leo Barcellos retweeted
How it started: Swarms, Orchestrators, Agent meshes, Multi-agent frameworks, Hierarchical agent systems, Agent-to-agent protocols How it's going:
57
99
20
1,311
76,707
Leo Barcellos retweeted
Claude Code - Official Multi Agent Orchestration This is a disabled feature, but through CC Mirror, I have been able to enable it It is so simple, yet powerful and IMO beats every orchestration I've seen There's no point building your own Updating CC Mirror later today so that anyone can try this github.com/numman-ali/cc-mir…
53
72
10
789
114,459
[Q:] Hello. I'm Q, the creator of an agent harness called `oh-my-opencode`, an @opencode plugin. I packaged a setup I had been using personally into a plugin and released it, and in less than two weeks, it has garnered over 3.4k stars. I used to be an ordinary backend engineer doing mundane backend work. I still recall a conversation I had with an engineer from the AI team on my way home from work last year. "All this talk about 'agents' is just a scam to get funding." It really is. Surprisingly, it is. Yet, here I am—someone who now loves agents more than anyone and constantly thinks about how to use them effectively—having said exactly that just a year ago. The turning point for me was the hook feature in Claude Code. "I can actually control and automate agents exactly how I want!" Since then, I’ve spent half the year obsessed with this agent, coding and experimenting frantically the moment I got off work. Right from my room. I could go as far as to say I dedicated my life to it. We are currently in an era that is changing at an insanely rapid pace. Everything we thought we knew is crumbling, and new, unknown things are being discovered every single day. We need to use every last token to think fiercely and question everything. We must be prepared to forget what we knew yesterday—what we considered facts—by tomorrow. Perhaps by this time next year, we might see teams arguing, "At least one person on the team needs to understand the entire codebase!" (For the record, I failed my prediction last year, so I don't mind if this one fails too, haha.) We really don't know. Let's question everything we know and build from scratch. Let's shatter every premise and rethink it all. And let's stand together in this world as it transforms. I offer my infinite gratitude to Boris for creating the hackable agent loop that sparked all of this.
When I created Claude Code as a side project back in September 2024, I had no idea it would grow to be what it is today. It is humbling to see how Claude Code has become a core dev tool for so many engineers, how enthusiastic the community is, and how people are using it for all sorts of things from coding, to devops, to research, to non-technical use cases. This technology is alien and magical, and it makes it so much easier for people to build and create. Increasingly, code is no longer the bottleneck. A year ago, Claude struggled to generate bash commands without escaping issues. It worked for seconds or minutes at a time. We saw early signs that it may become broadly useful for coding one day. Fast forward to today. In the last thirty days, I landed 259 PRs -- 497 commits, 40k lines added, 38k lines removed. Every single line was written by Claude Code + Opus 4.5. Claude consistently runs for minutes, hours, and days at a time (using Stop hooks). Software engineering is changing, and we are entering a new period in coding history. And we're still just getting started..
9
11
2
430
53,623
Leo Barcellos retweeted
Hoje os devs estão caminhando para 3 tipos: Vibe coder: versão IA do dev low/no-code. AI-Augmented Developer: aquele que ainda está preso na codificação mas já usa IA pra ajudar. Bem útil em alguns tipos de projetos específicos. AI-Driven Systems Engineer: Dev que entendeu que IA chegou e vai mudar a profissão e usa IA mais profunda para codificar.
12
5
3
126
12,753
Leo Barcellos retweeted
I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last ~year and a failure to claim the boost feels decidedly like skill issue. There's a new programmable layer of abstraction to master (in addition to the usual layers below) involving agents, subagents, their prompts, contexts, memory, modes, permissions, tools, plugins, skills, hooks, MCP, LSP, slash commands, workflows, IDE integrations, and a need to build an all-encompassing mental model for strengths and pitfalls of fundamentally stochastic, fallible, unintelligible and changing entities suddenly intermingled with what used to be good old fashioned engineering. Clearly some powerful alien tool was handed around except it comes with no manual and everyone has to figure out how to hold it and operate it, while the resulting magnitude 9 earthquake is rocking the profession. Roll up your sleeves to not fall behind.
2,554
7,298
2,906
55,177
17,120,846
Leo Barcellos retweeted
𝗦𝗮𝘆 𝗵𝗲𝗹𝗹𝗼 𝘁𝗼 𝗲𝗻𝗱𝗹𝗲𝘀𝘀 𝗰𝗿𝗲𝗮𝘁𝗶𝘃𝗲 𝗽𝗼𝘀𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝘆! 👋 🍌✨ For marketers, the creative process just got a massive performance boost. Our new, most powerful image model—Nano Banana Pro (Gemini 3 Pro Image)—is bringing cutting-edge AI editing capabilities right into your workflow. Want to master the art of prompting this powerful tool? Read this marketer’s guide to start transforming your campaigns today directly in Google Ads Asset Studio.
16
113
4
946
66,555
Leo Barcellos retweeted
"I'm a full-stack developer" the stack:
73
255
16
4,967
70,095