opencode agent create Go ahead, give it a shot.

I don't like ads. You don't like ads. We don't like ads. Limpa is like AdBlock + SponsorBlock but for podcasts. GitHub repo below open.substack.com/pub/duarteo...

Running a web server + a worker used to be a pain. Thanks Claude, didn't know about this one. github.com/nickstennin...

Now with agents you can have super meaningful commit messages. It's incredible. Lul

Django background tasks feedback. Me like.

complexipy kreuzberg throttled-py batchata I'm living under a rock? Never heard of any of these. Tasty though. tryolabs.com/blog/top-py...

Dirty little secret. Cancel Claude code, cancel ChatGPT plus. Github copilot is 10 bucks a month. And OpenCode can use them all 🤫

2000 km for 2025 ✅

Github actions in a nutshell: Comment in an issue -> get a notification for a failed action 😂

@reckless.bsky.social @davidpierce.xyz lightning round material? screenrant.com/stranger-thi...

2 themes built 1 attempt at building an lsp for beancount account completion 34 completion engines later. Done. Now we can carry on. github.com/duarteocarm...

Nothing like the fresh smell of a new Django project.

Part of my move away from LazyVim. My config is now 200 lines of pure lua and I love it. github.com/duarteocarm...

This month in Retrospectiva: - Neovim - OpenCode - A kitchen gadget that changed my life - A new health tracker I'm in love with - And some music. Last post of 2025! Enjoy the holidays :) open.substack.com/pub/duarteo...

We're definitely fighting the fight. The amount of trash out there hasn't stopped increasing. As always, the less code the better. www.youtube.com/watch?v=Ioi...

I have to say I'm a bit sick of LazyVim. I'm moving to NeoVim's default package manager: Pack. Fast, simple. Not finished, but here it is: github.com/duarteocarm...

In this post I dive deep into using LLMs for nutritional content analysis of meal descriptions. We talk DSPy, GEPA, on-device inference, and more! duarteocarmo.com/blog/from-n...

Publishing an app in the app store is such a broken process. - Builds, versions - App store reviews (sometimes 30 mins, sometimes 4 days) - Publish to app store connect takes 1 hour Is there a (SIMPLE) CI/CD, push, increment app version publish (SwiftUI)

thinking no-thinking reasoning_effort max_reasoning_tokens thinking_enabled include_reasoning reasoning_summaries It's just me? Or are y'all also a bit sick of this 😂

Do I have to come here everytime a new model comes out?

When do they send the badge? @OpenRouterAI

A good rule of thumb. Have an eval script ready for your application. New model comes out? Add the string and run. Will answer the "Should we be using this model?" real quick. Measure: - Time to first token - Latency - Verbosity - Quality (llm as judge)

Alright @opencode I'm sold.

Pro tip: You have multiple lambda intertwined, it's a mess. Direct them all into the same log group and run: aws logs tail /aws/lambda/yourlambdaname --follow Now everything's in one place

The cost of writing code definitly has gone to zero. The cost of writing quality code though. That one seems to be going up 😂

We are all a bit spoiled no? Claude code credits done? Opencode with github copilot time? Done? Codex it is. Done? While everyone is building the best agent possible, we're just a bit spoiled 😂

If I told you how many vibes I've cleaned up this week. You'd be amazed.

"Mathematician extraordinaire" Or, I probably shouldn't use @raycast to do basic math so much 😂

Denmark has a map of all the different AI projects being developed in the different municipalities It's crazy to me sometimes how fast this country moves 😂 www.kl.dk/videncenter...

Should rethink my branch names?

First day of sun in 43 days in Copenhagen. ☀️ Everyone is just instantly in a good mood Hurry up southern embassies. Send us those passports!!!!!

When the LLM actually *finds* the right function inside of an obscure library that I didn't know about. Good LLM. That almost looked like a hallucination.

Can relate

LLM app for mac: - chatgpt: limited - Bolt: buggy - Msty: buggy - Ollama: limited My new favorite: Cherry AI - Chinese open source for the win again https://www.cherry-ai.com

I'm still on the lookout for the best local LLM app on my mac. Hit me up if you have suggestions.

Why do you have a "legacy" section in your README? ARE YOU NOT AWARE OF VERSION CONTROL!?

This month in Retrospectiva: newcomers, OpenMeet, Marimo, LLM pre-training, Duruks, and Spanish thrillers. Enjoy! duarteocarmo.com/blog/retros...

Back in May this year, Alex reached out to me and asked me if wanted to collaborate with him on a new book. 6 months later, I'm happy to announce that DeepSeek in Practice is available for pre-order! More info about the release and links inside duarteocarmo.com/blog/book-r...

Updated my rss-to-podcast engine. And I'm sorry to admit. That TTS, still sucks. duarteocarmo.com/blog/tts-st...

Retrospectiva is a monthly newsletter about life and what I've been up to. Enjoy! duarteocarmo.com/blog/retros...

Faísca (spark in Portuguese) is a single 1000 line of code uv script that implements pre-training, supervised fine-tuning and GRPO from scratch - in PyTorch. The only thing you need to run it is uv. (GitHub repo inside) duarteocarmo.com/blog/faisca...

A couple of weeks ago I wrote a short article about AI and the utility bubble. Today, it got published on the Portuguese newspaper I read everyday! www.publico.pt/2025/10/08/... Google translate link in the comments.

More and more news. Thankfully, we still have open source. Quick post on how I ditched Feedly to regain control of my news feeds. duarteocarmo.com/blog/drowni...

After talking to a lot of "non-ai" people this summer. I was surprised at some of the misconceptions they had about AI/LLMs. This post captures just a few. open.substack.com/pub/duarteo...

In "RAG tricks from the trenches" I cover: starting simple, embedding choices, leveraging structured outputs, and keeping your evals pragmatic. Big thanks to the amazing staff and volunteers for (another!) amazing conference. Video: www.youtube.com/watch?v=oKX...

I regret to inform I am now in the Obsidian train. Blog posts might follow.

I'm giving a workshop on the 16th of August about how building agents backed by DeepSeek. We'll cover tools, MCP, and a couple of other interesting bits! www.eventbrite.com/e/deepseek-...

I spent the past couple of weeks adding a benchmark for european portuguese to EuroEval. Here's a small write-up and some preliminary benchmarks! duarteocarmo.com/blog/a-bench...

... but they can also be a whole lot of fun. 200 lines of Python, a uv script - and we are off to the races (literally)

How I used evals to systematically improve my LLM-powered food tracking app Taralli. open.substack.com/pub/duarteoc...