Innovating on what software collaboration and understanding could be! Ex-Google, Ex-Airbnb

Joined June 2025
Starting a new open source project with the aim of developing an artifact standard for verifiable system diagrams. Check out some of the demo diagrams principal-ai.github.io/subsy…
2
1
1
14
1,827
Doing an analysis of the constructs for the human layer repo
1
86
Fernando retweeted
i think i kept seeing people say "monorepos are good for agents" but we just switched to multi repos. i have a hard time caring about the benefits of monorepos anymore
227
32
28
1,873
191,642
Fernando retweeted
I've been working on a new visualisation experience for users in @event_catalog . With "Levels" you can now decide what view of your diagram you wish to see. Level 1 - Domains, systems and actors Level 2 - Services and data Level 3 - Messages and channels These new levels can help your teams understand your architecture through different lenses either staying at a high level or diving deeper. Now live 🙌
2
6
86
4,646
We've been working on building trace intelligence across all your company sessions. We're building dimensions of trace data especially relevant to the offices of a CTO, CISO, and CFO. You'll be able to answer things like "how often do files in your codebase show up in agent read operations and how much does it cost" (an example visual below) This is a fun but gnarly data engineering challenge. But IMO the real story is all the software we are creating along the way. We're building internal tools to evaluate schema choices, we use data from our agent to evaluate the types of queries to optimize for (using Jev and Luna to classify), we build visual profiling tools test query performance throughout. Making software now is so much more fun and interesting. Anyways, I can't wait to show this stuff off. Your agent trace data contains multitudes and you need many ways to look at it. Also, if this is interesting enough that you might want to run a pilot, please DM!
3
2
4
37
1,644
Fernando retweeted
I spent the majority of my career building observability tooling this tooling should be used to close the loop between production and coding agents, in service of self-improving software I put my thoughts down boristane.com/blog/self-impr…
39
49
3
415
20,744
Fernando retweeted
People ask me how I ship omarchy sw so fast. 🧵 The answer is Clarity. Clarity is where all my projects live, from the first idea to running live. Every project gets a map, a status, its questions and decisions, a backlog, testing and its GitHub, in one place. Every AI chat I open reads from it before it starts. Made with Clarity with a buddy of mine, lets call him Daniel, not on X and wont be. Here's Blip, my iMessage app for Linux, going through it.
27
13
3
319
26,999
Fernando retweeted
Code is cheap now. Understanding isn't. Introducing Spinal: a low-latency, predictive surface to grok and keep up with any codebase. From weekend hacks to mature behemoths. Linux kernel, 1.48M commits: first search results in <20 ms. Waitlist: spinal.sh
17
19
5
433
32,746
Fernando retweeted
👀👀 "[…] we provide several examples of AI mistranslations of natural language (NL) statements and proofs into Lean in practice, resulting in mismatches between NL proofs and their Lean 'verifications'. These include OpenAI's announced Navier-Stokes proof. In particular, we show that the formalised Lean proof does not correspond to the NL proof of blow-up of solutions to the Navier-Stokes equations." arxiv.org/abs/2610.08144
12
96
31
691
104,994
Fernando retweeted
PR guides are now available in capy and they have made reviewing code so much easier - overview summary with diagram breakdown - diffs broken down and sorted into logical sections - tests and generated files collapsed by default - moved code detection
Plugins and guided reviews are here! Capy 0.4.4 • Plugins from the Capy marketplace • A guide on every reviewed PR • Custom models on your own key • Desktop threads without a project • Fixes and improvements capy.ai/changelog/0.4.4
2
8
59
3,440
Fernando retweeted
working at a higher level of abstraction has always required understanding the lower level ones, I don't think coding agents changes this
301
415
89
5,623
359,271
This will replace Excalidraw for me. Ferroprint. Ferroprint after the ferroprussiate process that made the first blueprints
I love blueprint for some reason
52
125
6
1,744
64,784
I built Petal, a free, open-source disk-space explorer for macOS. The chart draws itself while it scans. It flags what's safe to delete (Xcode builds, node_modules, caches) and tells you exactly what deleting frees. Rust + GPUI. Inspired by the recent syntax.fm episode. github.com/henrydennis/petal
30
20
1
363
22,693
maybe don't ship 2k+ file PRs But if you do, let the PR explain itself Try it here: github.com/coldteadotai/pr-l…
A pull request with 2,200 files, 1M+ changed lines, and 400+ inline comments. Impossible to review? Not if you have the GitHub Copilot app. ✨ The rebuilt pull request view keeps even this extreme diff fast and smooth. 🧵
7
2
54
13,600
Fernando retweeted
If you get bored drawing architecture diagrams, you can now spawn a rider and drive around your diagram. Draw and press SHIFT + R on koboyo.com
9
18
4
213
17,165
Fernando retweeted
Software factories are an anti pattern. Agents are incredibly useful but you must still be deeply engaged to build a great product.
17
2
30
2,839
Fernando retweeted
had a crazy idea today for the past two months i've been building a pretty massive app and truth be told, i often play catch-up on architecture, because features move so damn fast so today i asked claude (opus 5.5, high) "I have a weird task for you. I want you to look at the app and create, for lack of a better word, a canvas that I can zoom in and zoom out and move around to see how our architecture actually works." (full prompt in first comment) this came out pretty great and it truly visualized, with steps and connected graphs what everything does 10/10 would recommend
88
92
10
1,655
190,357
The hardest part of software engineering was never how do you write code that does what you want, but how do you know that your mental model is accurate. This is the problem other tools fail to solve, surfacing what happens and where so you have confidence in your system
1
4
275
Fernando retweeted
Replying to @thdxr
This isnt a motion one, but it was solid for just "make a release video for this project"
1
261