- 0
- Followers
- 0
- Median views
- 0.0%
- Engagement
- 0
- Views this week
This week on X
Oct 27 – Nov 2 · views on the posts made each day, replies left outRecent top performing posts
Live from X> AI bubble will pop soon > none of these companies are profitable > AI labs are burning billions for nothing > it's all hype and no substance > AI is just a glorified autocomplete > AI progress has hit a wall > it's just another dot-com bubble crowd is about to be in shambles
Yann LeCun says: People in AI have made the same mistake for 70 years: if a machine can do a task humans see as intellectually complex, it must be smarter than humans. This is completely false We have computers that are superhuman at many tasks but we're still far from machines

absolutely brutal results for open-weight models on the updated HLE Diamond: > chinese models are over 30 points behind > even meta muse wonder is 20 points behind gemini 4 carbon is the real deal along with opus 5.5 but kimi k3.1 which claimed to be near fable 5.1's level is
Who engages with Haider
Real people replying on XWhat they do
306real people replying to recent posts74 with 1K+ followers
Engineers42
Founders39
Creators26
Researchers21
AI engineers18
AI founders17
2 of them work at
Stanford University
From each person’s bio on X.
Where they are
136 with a locationVerified on X
- Verified59.5%
- Not verified40.5%
When they reply
434 replies · UTCBusiest: Wednesday, 3pm–6pm
Their own reach
Followers each of them hasRecent collabs
5 brands · paid partnerships in their postswe've gotten really good at making voice agents sound human but i think we're still missing something more important: understanding the human on the other side @hume_ai is launching its Decision API which analyzes audio directly and lets developers ask questions about tone, emot
849one of the more interesting things happening in AI right now is the shift from personal assistants to agents that can actually work together @zenzap lets businesses create AI agents, connect them to existing tools and automate workflows
2.4Ki like @digitalocean's new agent stack especially Simulations being able to test agents against real production scenarios, evaluate entire conversations and mock expensive tool calls before shipping changes is a big deal could be huge for production agents
5.2Ki think building capable agents is getting easier than making them work reliably in real enterprise systems every customer has different objects, fields, permissions, and workflows Ampersand handles that integration complexity, including authentication, retries, and monitoring
2.4Kone of the harder parts of building agents is knowing if they actually work outside a clean demo Era by Eon generates a synthetic company with connected CRM data, tickets, Slack, files, calls, and more, so you can test how agents handle real workflows across systems i think thi
2.6KLaunch with Haider
Send a brief, approve the draft, pay once it’s live.
Single post
Quote post
