4x CTO - Now CTO of @magmapartners, prev @ycombinator w21 AI & VC Enthusiast. 🤹🏻♂️ Juggler 💻 Hackaton's winner ✈️ traveler 🤖 Maker
Joined April 2009
- Tweets 5,644
- Following 3,556
- Followers 1,010
- Likes 25,076
507 Photos and videos
May 20
I've come to the conclusion that @Railway must have its infrastructure in the Yucatan Peninsula because the power grid just went out and my sites stopped working.
43
Apr 11
Si trabajas en logística, esto te interesa.
Empacaste mal → pagaste por mover aire.
Los algoritmos de packing 3D existen para que eso no pase.
Escribí sobre los principales: 3D-FFD, 3D-BFD, layering y algoritmos genéticos.
tienen demo funcional👇
1
1
2
72
Apr 11
Te dejo un artículo con los detalles de mi investigación dbug.mx/blog/algoritmos-empa…
1
1
38
Iddar Olivares retweeted
Apr 8
Tengo 35 años y cancer de mama metastásico, un caso raro, menos del 1% de tumores de mama son como el mío y hay poca documentación sobre ello.
Por eso me gustaría encontrar personas que se dediquen a esto y que quieran investigar con mi caso. Twitter haz tu magia
734
16,288
21,607
1,419,179
Mar 22
🚀 Estamos contratando en CDMX
Busco 2 perfiles Sr:
📊 Científico de Datos
🔧 Ingeniero de Datos
✅ Python SQL
✅ Pipelines y automatización
✅ Experiencia en retail o consumo masivo
✅ Contratación inmediata
CV a 👉 iddar@dbug.mx
RT apreciado 🙏
3
26
68
6,183
Mar 22
Si conoces a alguien que trabaje con volúmenes grandes de datos, tenga ojo para el negocio y ganas de construir algo serio — ponme en contacto.
DM abierto 📩
1
555
Iddar Olivares retweeted
Feb 18
Today I'm launching Interpreter.
It's a desktop agent that can fill PDFs, edit your Excel and Word docs, and learn new skills.
Runs offline, works with any model, and it's free.
221
326
4,604
365,788
Iddar Olivares retweeted
Jan 29
Introducing LlamaBarn — a tiny macOS menu bar app for running local LLMs
Open source, built on llama.cpp
ALT LlamaBarn
22
63
779
54,189
Iddar Olivares retweeted
Jan 20
🦀 Rust's standard library on the GPU 🤯
VectorWare shows how they enable devs to write complex, high-performance applications that leverage the full power of GPU hardware using familiar Rust abstractions without #![no_std] pain:
vectorware.com/blog/rust-std…
#rust #rustlang #gpu
7
74
558
27,509
23 Sep 2025
Building with LLMs is exciting… but there are 3 hidden headaches most teams face:
•Versioning prompts
•Evaluating quality over time
•A/B testing across models/providers
1
171
23 Sep 2025
How do you measure LLM quality?
•Accuracy isn’t enough
•Human eval doesn’t scale
•Metrics drift as data changes
Long-term quality tracking is still an open challenge.
2
142
23 Sep 2025
Curious to hear from others:
👉 How are you handling prompt versioning, eval, and A/B testing today?
Tools? Frameworks? Homegrown hacks?
Let’s share what’s working 👇
190
23 Sep 2025
Do you A/B test prompts or models?
Switching between OpenAI, Anthropic, Qwen, etc. is powerful—
…but without consistent eval frameworks, comparing apples to apples is tricky.
101
14 Sep 2025
I’ve been an Arch Linux user for years.
But even after a clean install, I still had issues with Bluetooth & peripherals.
1
122
14 Sep 2025
Minimalist design, smooth experience, and just a couple keymap tweaks.
This is the Linux experience I was waiting for.
1
96
12 Sep 2025
Big thanks to @charliermarsh and the @astral_sh for building UV 🙌
Testing Python apps across different OSs has never been smoother.
Fast, simple, and finally a headache-free way to manage Python envs.
A true replacement for Conda/Virtualenv. Amazing work
1
6
613