David Ondrej
@DavidOndrej1
building http://vectal.ai
cursor + claude code + superwhisper + o3 pro + vectal guys, i think i'm about to ascend i never thought i'd reach this level of productivity i'm shredding through github issues at an insane rate...
Kimi K2 is one of the best models for AI agents—it's affordable, reliable, and highly capable. Yet, most people are using it incorrectly. That's exactly why I decided to build these four AI agents: to showcase its true strengths.
i love how everyone understands what "pavel durov'd myself" means
Anthropic has the WORST deep research feature in the game. > checks the most sources > results absolutely suck better off just doing a simple Perplexity query...
Claude Code has been on fire today... what is going on!! the tool is fucking amazing, i'm getting so much done
I love Claude Code as much as a man can love software.
the only real education about LLMs is getting your hands dirty. you're going to learn 100x more trying to build your own LLM-powered app, than reading all the books about this subject that exist.
if you see this, and you don’t drop everything to work on AI… i don’t know what to tell you
AI TL;DR: 1. OpenAI gets gold medal on IMO 2025 2. Googlegets gold medal on IMO 2025 3. Kimi K2 is the best open source model ever 4. Qwen outperforms it All within a single week.
if this causes you nightmares... you're ngmi
Companies are using fake humans and AI to do interviews now...
all the CLI tools absolutely SUCK when compared to Claude Code. the UI and UX of Claude Code is simply unmatched.
the fact that Europe gets AI features later than Afghanistan is mind-bending. we live in a fucked up world.
Agent mode is now available in Europe! 🇪🇺🇪🇺
models like Kimi, DeepSeek and Qwen will cost the closed AI labs BILLIONS of dollars. that's why nobody is talking about them. despite these LLMs absolutely crushing all of the benchmarks. Claude 4 Opus is literally *100x* more expensive than Kimi K2 yet both models have…