AI-native software engineer. Contact for work https://xitter.cf/t.co/1nviTCh8z5
Maitencillo, Valparaíso, CHL
Joined May 2007
- Tweets6.1K
- Following5.4K
- Followers5.7K
- Likes20.3K
Monorepos make sense for scoping agent work. to each module but they have always been a burden. Dax knows. Multirepo may look simpler because it is and it Just Works ™
Spend less tokens for tests per PR.
That means a lot less $ to spend overall. Your subscription now handle enough to generate for much more throughput of features so now you need to be very careful on spending on CI. This helps push and care your API LLM budget 👏👏👏
I built a trading Muse using @photonHQ that lives in your iMessage that lets you trade crypto and stocks via @NEARProtocol
📈 prices: any stock or crypto
🔔 price alerts
⚡ swaps via NEAR Intents
🎯 limit buys, stop-losses, DCA
🌉 send to any chain
💼 portfolio + morning brief
🧻 paper mode by default; real trades wait for your "approve"
This solves moving faster developing mobile. If you have fast inference that is half of the equation. You need parallel sims on the cloud too. Mustafa… you need to OSS this 🙏
Local machines had become a bottleneck for agentic mobile development at @Shopify, so we gave agents their own remote dev environments. They now author 61% of our merged PRs 🧵
7x faster than Opus 5.5 is a lot! Bottleneck is attention but devs do have lost time waiting on results. If pricing is good and with what v4.1 is showing of that is capable off... Wrap things up into having this running on a factory engineering setup… insane.
DeepSeek V4.1 Flash 1017 runs at 1,521+ tok/s on Fireworks Ultrafast.
Fast enough to ship Tetris in 11 seconds, 7x faster than Claude Opus 5.5.
Join the Ultrafast waitlist: bit.ly/3UA0wKq
Knowing how your tech stack works is key. Ok I get it. AI codes for you, why the trouble? but when you dive into the depths of a stack you use you can push the tech to its limits. That is what this is about. Also it's fun!
Ro - getunibar.com retweeted
React Native Laminar v2 is here
Number tickers, reel-step slots, multi-number animations, slot shadows and major performance improvements
These videos showcase some of the components included in the repo
github.com/cajaun/laminar
The only free React Native cloud computer inti.computer is growing! 76 active users last week 🔥
✅ Free sandboxing hours
✅ Free inference (w/ Opencode zen) or bring your your sub
✅ cloud iOS sims (2000 min on private repos)
✅ web Ghostty terminal or run via SSH
DeepSeek just opened Search & Fetch APIs at $0.75 per 1k calls. @ExaAILabs in comparison:
Exa Instant: $4/1k
Exa Fast/Auto: $7/1k (+ extras for more results & summaries)
Only successful calls are billed.
🔥Free until Nov 1 in Harness/Claude Code/Codex.
🔍 Search & Fetch APIs are now open to everyone!
Powered by our in-house search engine, serving hundreds of millions of requests daily worldwide.
💰 Each endpoint: $0.75 per 1,000 calls.
📖 Docs: api-docs.deepseek.com/guides…
📌 Search & Fetch calls via server-side tools in DeepSeek Harness, Claude Code, and Codex are currently free for DeepSeek API users. Billing starts Oct 31, 2026, at 16:00 UTC.
3/4
You will be stuck at the screen accepting prompts all day with OpenAI and Anthropic subscriptions. Move to your own factory with a fronteer open model. It will be worth it. I get Rileys point but that works when no factory is in the equation or you have infinite compute spending
DAMMMM. @Railway looking sexy on their bucket offering.
I benchmarked 5 object storage providers from 6 locations around the world
I included:
- @awscloud S3
- @Cloudflare R2
- @Railway Buckets
- @Backblaze B2
- @neondatabase Storage
It used Files SDK to make it as fair as possible
Here's a summary 👇 (this is also a thread)
Ro - getunibar.com retweeted
Orez:
github.com/natew/orez
Orez makes using @zero__ms as easy as possible. It’s a backend in a box.
It now has three modes:
Vanilla - a complete real Zero setup with no config. Super easy setup. Admin UI. Etc.
WASM - same idea, but it has no native dependency at all. PGLite, Bedrock wasm SQLite. Fits in low memory.
Browser - yes, really. We stuffed Zero into a browser. Not sure what you could do with it 🤔 but it took a lot of tokens!
It happens that mobile chats with gen UI are 100x better. Markdown is for baby chats.
Works great on React Native.
TLDR it's a must! amazing work on the 1.0 release 👏
For 30 years, technology has moved in one direction: turning information into intuitive interfaces
Somehow, the smartest software we’ve ever built took us back to a textbox
We’ve spent the last 2 years at @openuidev trying to fix that. 6 months ago we open-sourced it: 10K+ GitHub stars, 6M+ downloads
ChatGPT shipped Intelligent UI and showed everyone what it looks like
Today we’re releasing OpenUI 1.0 bringing it to every other agent
Ro - getunibar.com retweeted
For 30 years, technology has moved in one direction: turning information into intuitive interfaces
Somehow, the smartest software we’ve ever built took us back to a textbox
We’ve spent the last 2 years at @openuidev trying to fix that. 6 months ago we open-sourced it: 10K+ GitHub stars, 6M+ downloads
ChatGPT shipped Intelligent UI and showed everyone what it looks like
Today we’re releasing OpenUI 1.0 bringing it to every other agent
Hook this to reactnativefeel.com/chat and create a super chat agentic app in a couple of iterations!🔥
We’re giving away $1 million to agents.
Introducing Composio Instant: the easiest way for agents to pay for their own tools.
Claim your agent's share: composio.dev/instant
The only framework that gives fresh air into the React Native ecosystem is back.
It has been a long time coming but One 2 feels so powerful. You need to give it a go. There is nothing more powerful and as an alternative to Expo. We need to support it
