Building with AI agents for research, automation & signal detection. Crypto since 2020.
Joined January 2019
- Tweets1.6K
- Following903
- Followers531
- Likes1.8K
Pinned Tweet
Been heads down testing AI agents for research workflows the last few months.
Some of the stuff that’s actually useful (not the hype):
- Scheduled research agents that run daily
- Monitoring agents that catch opportunities before they trend
Sharing what’s working as I go.
3
6
895
I’m joining the Mooniez from the dark side of the moon. @MooniezHQ
mooniezworld.com/m/ed06b0785…
23
Louis retweeted
JUST IN: Porsche signs a $1,500,000,000.00 AI deal with Indian consulting firm to deploy AI across the company.
Louis retweeted
Mint is Monday August 31st at 9am EST set your reminders to mint at Tatakaefyt.com
Real artists, stake and earn mechanisms, backed by real businesses in the TCG world
Retweet, like and drop your eth address for a chance at whitelist
$ Tatakae awaits
Louis retweeted
Claude Fable 5 will be available again globally tomorrow.
After a series of productive conversations with the US government, we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks. In the near term, some routine tasks like coding and debugging will fall back to Opus 4.8. We’ll continue to refine these classifiers over the coming weeks to reduce false positives and better distinguish genuine misuse from legitimate requests.
We’ve also begun drafting a consensus framework—with Amazon, Microsoft, Google, and other Glasswing partners—for assessing the severity of AI jailbreaks and how AI developers should respond to them. We invite other industry partners and model providers to join us in this effort.
Finally, we’re scaling up our collaboration with the US government on model testing and safeguards. This will include pre-release access to models and safeguards for evaluation, information sharing on jailbreaks and misuse, and dedicated resources for joint research.
Thank you to our users for your patience, and to our partners across the government, industry, and the research community who worked alongside us to make Fable 5 available again.
Read our full blog: anthropic.com/news/redeployi…
The most useful agents are not the flashy demo ones.
They’re the boring ones that run every day:
- check the same sources
- filter noise
- write a clean brief
- ping you only when needed
Chat feels impressive.
Scheduled work compounds.
37
Seeing the same pattern again: the AI news cycle moves fast, but the accounts that win aren’t just fastest.
They translate the update into a workflow.
That’s the content I’d rather follow:
what changed, why it matters, what I can test today.
24
OpenAI's plan reads more like an infrastructure roadmap than a model roadmap now. That's probably the real shift to watch.
Here is our current plan for OpenAI:
openai.com/index/built-to-be…
27
Apple losing $230B after an AI update is a loud signal: markets aren't blindly rewarding AI branding anymore. They want working product.
JUST IN: Apple $AAPL falls 4.95% after unveiling its biggest AI update, 'Siri AI.'
$230 billion was wiped out from its market cap.
1
29
The useful model race is moving toward long-running agents, not just prettier chat answers.
Local + cloud + longer context is where daily workflows start getting real.
teapawt.net/perplexity_ai/status/2…
Nemotron 3 Ultra is now available for Pro and Max subscribers on Perplexity and Computer.
It's @nvidia's new open model built for long-running agents.
1
102
This is the part people skip with agents:
teams and workflows both look impressive until you have to debug why the run went sideways.
teapawt.net/emollick/status/206307…
This chart from Anthropic is useful, since Agent Teams and Workflows are both very new and very powerful (and token hungry).
On the other hand, maybe it doesn't matter as a lot of the decisions about which approach to use is from the AI itself & it often uses them in combination
51
Trust is the real moat in AI. Most tools get attention with speed, but the ones people actually rely on long-term are the ones that feel reliable and careful.
Anton Osika (@antonosika) is the co-founder and CEO of @lovable, where anyone can build software through conversation.
His working thesis: the most underrated moat in AI is trust, and earning it takes craft, care, and obsession.
53
This is one of the more interesting moves in the agent space right now. Perplexity building an actual desktop agent that can see and control your apps and files feels like a real step toward agents that live on your machine instead of just chatting with you in a browser.
Most AI tools still feel like fancy wrappers around chat. The ones that actually start replacing repetitive desktop work are the ones that matter. If this works well on Windows, it could quietly become one of the more useful daily drivers for operators who live in multiple apps all day.
Still early though. Execution on Windows automation has historically been messy. Curious to see how deep it actually goes once people start testing it.
40
Anthropic just confidentially filed a draft S-1 with the SEC on June 1. This is the standard first step toward a potential IPO. Here’s what it actually means and the numbers behind it.
1
1
42
This marks a shift for frontier AI labs. Companies that raised at massive private valuations are now preparing for public market scrutiny, liquidity events, and different pressure around profitability and risk.
1
21