Alan.Dev
@AlanAcostaDev
Followers
168
Following
492
Media
18
Statuses
238
Digital Builder | Creator of @CodenceLab
Argentina
Joined December 2021
it's ridiculous how fast @opencode grew to >80K stars on github it took us 5 years at supabase. they did it in 8 months
65
53
2K
¡Acaba de salir el Figma de Claude Code! Se llama Pencil: ✓ Diseña, versiona y genera código ✓ Con servidores MCP integrados ✓ Funciona con VSCode, Cursor ✓ Para macOS y Linux Lo mejor, sin coste ni suscripción: → https://t.co/88IkfyDmPS
74
471
5K
The natural UI for skills? A skill tree 🌳
195
282
4K
Claude can now securely connect to your health data. Four new integrations are now available in beta: Apple Health (iOS), Health Connect (Android), HealthEx, and Function Health.
392
564
10K
Google just released its own MCP (Model Control Plane) server for Stitch, turning LLM coding agents into UI-savvy collaborators. The new Stitch MCP Server pipes designs directly into your IDE. 🔹 Generate new screens 🔹 Fetch code from any design 🔹 Give your agent full visual
Developer Week Day 2. 🛠️ We just gave your Coding Agent a Design Degree🎓🎨 Introducing the Stitch MCP Server. 🔌 You can now pipe Stitch designs directly into your favorite tools like Antigravity. * Generate new screens without leaving your IDE * Fetch the code from any
41
90
959
Running GLM-4.7-Flash on 4 x M4 Pro Mac Minis using @exolabs. Uses tensor parallelism with RDMA over Thunderbolt & MLX backend (h/t @awnihannun). Runs at 100 tok/sec. We're working on optimizing this at @exolabs. Aiming to hit ~200 tok/sec on this setup soon.
Running LLM evals with @exolabs on 8 x M4 Pro Mac Minis. M5 Pro Mac Mini memory bandwidth should be ~30% faster (~350GB/s) + 4x FLOPS w/ Neural Accelerators (tensor cores). 8 x M5 Pro Mac Mini cluster would have a total of 2.8TB/s memory bandwidth with RDMA over Thunderbolt 5.
41
67
586
Tests you can read and understand. Plain English. AUTOMATE your mobile app with Maestro. 🙂
10
23
355
Exploring this great tool created by @sabosugi. It aligns right with my personal and agency brands. ⚡
1
0
2
This is crazy, but it was expected.
We built a browser with GPT-5.2 in Cursor. It ran uninterrupted for one week. It's 3M+ lines of code across thousands of files. The rendering engine is from-scratch in Rust with HTML parsing, CSS cascade, layout, text shaping, paint, and a custom JS VM. It *kind of* works! It
0
0
0
Designers… how did we all miss this Tools You Need To Know About - (Day 82)
9
16
450
NOTICIA DE ÚLTIMA HORA 🚨: Meta estrena 'Meta Superintelligence Labs' (MSL) con Alexandr Wang a la cabeza. ¿Su primera medida de "eficiencia"? Wang sustituyó a toda la cúpula ejecutiva por versiones 100% optimizadas de él mismo mediante Claude Cowork.
20
20
351
Actualmente hay alguna mejor opción que TaskMasterAi ? ( TaskMasterAi sirve para que el ai assistant no pierda contexto del proyecto y empiece a cometer errores y/o alucinaciones )
0
0
1
Today we’re announcing Open Responses: an open-source spec for building multi-provider, interoperable LLM interfaces built on top of the original OpenAI Responses API. ✅ Multi-provider by default ✅ Useful for real-world workflows ✅ Extensible without fragmentation Build
131
482
4K
Dither / ASCII Effect Pro Free to Use: https://t.co/tF4SOoNUq3
#effect #ascii #dither #javascript #trends
23
90
1K
Meet the new look of @neondatabase. A total of ~1,200 hours of teamwork and five months from the initial kickoff to bring this design beauty to prod. Rebranding, design, dev, and motion. Ship faster with Postgres!
65
29
595
He estado trabajando con Claude Cowork los últimos 2 días, y SIN exagerar ha podido hacer el trabajo administrativo / back office que a un founder le puede llevar 1 semana. 1. Investor reports 2. Análisis de clientes (Conectando al CRM y Slack) 3. Job descriptions El futuro del
14
29
440