steve ike
@steve_ike_
Followers
1K
Following
3K
Media
259
Statuses
7K
AI, Tech & Design enthusiast. Building @magicllamaAI and @emx_dev.
✨🦙🤖👷♂️
Joined August 2024
Most people think AI progress is about smarter models. Sam view is sharper: models matter, but the real leverage is product, memory, and compute.
Sam Altman (@sama) just gave the clearest signal yet about where AI is actually headed. And it’s not the model race everyone is watching. — from pod with @Kantrowitz 1/ Most people think AI progress is about smarter models. Sam Altman’s view is sharper: models matter, but the
0
0
0
I tested all of the best vibe coding tools so you don't have to https://t.co/M2ibf3wR6y
read.technically.dev
Comparing Replit, v0, Lovable and Bolt, in a bakeoff to decide who will be Vandalay Industries go-to vibe coding tool.
1
4
49
Last week we added skills to Codex! Here are a few tips that I found helpful when working with skills 👇
6
20
206
Thomas Wolf (@Thom_Wolf) argues that jobs are not just tasks. They are made of three layers: execution, judgment, and agency. AI is rapidly automating execution, which explains why entry-level jobs are disappearing first, while senior roles remain resilient. The real bottleneck
I've been reading and thinking about jobs, economics and AI quite a lot in late 2025 Ended up writing some thoughts at https://t.co/AuVrbrIUxd First piece in a long time, should probably write more often
0
0
1
Shane Legg forecasts 50% probability of "minimal" AGI by 2028. https://t.co/MvFeafbUWd - Defines minimal AGI as AI handling typical human cognitive tasks in language, knowledge, and reasoning. - Predicts full AGI matching human cognition breadth within a decade, leading to
How do we build a world where AGI helps humanity flourish? 🌱 Our co-founder and Chief AGI Scientist @ShaneLegg discusses the potential for a "golden age" of discovery and the societal shifts we need to prepare for �� all on our podcast, with @fryrsquared ↓ Timecodes 00:00
4
14
71
Josh Woodward (@joshwoodward) has became one of the most powerful people in AI and most people have never heard of him. He runs the Gemini app, leads Google Labs, and is quietly reshaping how Google competes with OpenAI, without blowing up user trust. Here’s the story and why
1
0
0
Sam Altman (@sama) just gave the clearest signal yet about where AI is actually headed. And it’s not the model race everyone is watching. — from pod with @Kantrowitz 1/ Most people think AI progress is about smarter models. Sam Altman’s view is sharper: models matter, but the
0
0
0
Choice requires contrast. We only know what we like by preferring it over something else. Remove comparison and you remove confidence. — @rorysutherland on @farnamstreet/@shaneparrish pod
1
0
2
If you're one of these super-unicorns, you'll be able to write your own ticket. It's good to be you.
Because of LLMs, the only type of people I’d hire now are what I’d call “super-unicorns” - Developer - Designer (or great design taste) - Product-focused (vs feature-focused) - Marketer - Agent expert - High emotional intelligence - 5+ years experience creating web app products
22
7
208
As we think about intelligence and what qualifies as AGI, it really highlights how remarkable the human brain is.
Yann is just plain incorrect here, he’s confusing general intelligence with universal intelligence. Brains are the most exquisite and complex phenomena we know of in the universe (so far), and they are in fact extremely general. Obviously one can’t circumvent the no free lunch
5
0
4
Yann is just plain incorrect here, he’s confusing general intelligence with universal intelligence. Brains are the most exquisite and complex phenomena we know of in the universe (so far), and they are in fact extremely general. Obviously one can’t circumvent the no free lunch
Yann LeCun says there is no such thing as general intelligence Human intelligence is super-specialized for the physical world, and our feeling of generality is an illusion We only seem general because we can't imagine the problems we're blind to "the concept is complete BS"
714
1K
10K
NVIDIA just open-sourced a 30B model that beats GPT-OSS and Qwen3-30B - and runs 2.2–3.3× faster. Nemotron 3 Nano: Up to 1M-token context MoE: 31.6B total params / 3.6B active Best-in-class SWE-Bench performance Open weights + training recipe + redistributable datasets And yes:
36
98
937
When it comes to scientific discovery, one thing LLMs are really good at is getting hobbyists to delude themselves into believing they've made a huge breakthrough on some longstanding problem or a theory of everything
265
239
3K
If you want opportunity, leave breadcrumbs. Make one thing so good that people cannot ignore it. The internet will do the rest. — @patrick_oshag on @FoundersPodcast
1
0
9
There is always room for great.
Most people chase goals. The best builders chase principles. @FoundersPodcast conversation with @patrick_oshag completely reframed how to think about success, work, and a meaningful life. Here are the ideas I cannot stop thinking about: 1/ The most exciting thing in the world
0
0
1
From Karpathy’s latest post: “LLM apps will organize, finetune and actually animate teams of them into deployed professionals in specific verticals by supplying private data, sensors and actuators and feedback loops.” This is exactly right. In a world of AI agents, there is a
61
92
618
Most people chase goals. The best builders chase principles. @FoundersPodcast conversation with @patrick_oshag completely reframed how to think about success, work, and a meaningful life. Here are the ideas I cannot stop thinking about: 1/ The most exciting thing in the world
8
18
199