Diego Castillo

Collaborating on Markdown files in tools like @github.com is honestly painful. You can't comment on a specific line, can't suggest inline edits. They need to become first-class citizens. @notion.com handles this way better, but it's not where my code lives.

Claude Code had a very successful start to the year. The aftermath seems to be that throwing shade at it is now fashionable.

New favorite workflow: end the day with a solid implementation plan split into a few tasks, hand them to coding agents, wake up to multiple PRs ready for review. 🫡

The real bottleneck is clarity of thought.

GitHub stats from the last couple of months would be fascinating to see. Feels like the amount of code being written quietly exploded due to agents.

Playing with agent skills this week.

Which ones have actually stuck for you, and what do you use them for?

The models are fine. Figuring out what you want is the hard part.

If anything, CI and code review have become the bottlenecks. Those are two areas that will be fascinating to watch as agentic coding continues to gain traction.

The speed we can move as software engineers right now is honestly wild. I'm shipping more every day than I ever have throughout my career, in ways that simply weren't possible before. Interestingly, I also have more work than ever, not less. 😅

The 80/20 prompt I use with coding agents before creating the plan: "Ask questions if any."

One line. Massive difference.

I've used the Pro tier ($17/month), and when you take a bit of care to write thoughtful prompts, it works really well.

Back to Neovim + Claude Code.
All other subscriptions gone.
I wonder what this looks like in a year. 🤔

Life lately 😅

LLMs are the new level of abstraction.

Leverage them, or someone else will. 🤷‍♂️

Monday runs on coffee.

Special thanks to @marochko.bsky.social and @danielparas.bsky.social for helping us improve the implementation even further. 🙌

This feature will start rolling out to users soon. Looking forward to seeing it in the wild. 👀

I really like the idea behind CX Week. Ship something valuable, work with people you do not usually collaborate with, and show customers that we are listening. It is a great way to close out the year.

It took real async collaboration to pull it off, which made the experience even more rewarding. 🌎

I worked with Daisy and Ben on adding the ability to see previously used Threads topics in the Buffer composer. We finished it within the week, despite working across three continents, with teammates based in Thailand, Australia, and Panama and up to 16 hours between us.

This week at @buffer.com, we wrapped up Customer Experience Week ✨, a few days focused on making tangible improvements for customers and a nice way to end the year strong.

Kalita pour over, Pacamara character. Quiet joy.

Brewing a moment.

Daily pour over joy

Have you onboarded at a company recently? What interesting ways have you used AI during onboarding? I'd love to hear your thoughts and experiences. 👇

Onboarding with AI tools available has been smoother. I can ramp up faster, understand the codebase more quickly, and get unblocked more easily. The tools help me contribute sooner while still building the relationships that make remote, async work successful.

🧘 The Balance

It's tempting to work in a silo when you can unblock yourself, but that misses the point. I still reach out, ask questions, and pair on complex problems, just with clearer context and specific questions instead of broad confusion.

Plan mode has been useful here too. Before committing to a solution, I explore multiple approaches, weigh trade-offs, and make more informed decisions. I don't just follow what AI suggests. I use my own engineering experience to evaluate what it proposes and decide what makes sense.

⚒️ Getting Unblocked

Local setup issues are common when joining a new codebase. AI has been great at understanding error messages, suggesting fixes, and helping me get unblocked without waiting for someone else to be available. In an async environment, immediate help isn't always possible.

This extends beyond the codebase too. When I need to understand company workflows like retrieval, storage, or validation, I point AI to our docs so it can map out how things work internally. That way I can propose solutions that follow our patterns instead of starting from scratch.

github.com, Linear, and @notion.com agents have been particularly useful here. Instead of spending hours searching through repos and docs, I get pointed in the right direction quickly, then dive deeper.