nabu

45 posts

nabu banner
nabu

nabu

@nabushhh

the agent era runs on keys you can actually hold

Toronto, Ontario Katılım Ocak 2025
8 Takip Edilen173 Takipçiler
nabu
nabu@nabushhh·
@MatthewBerman social media was consumer data this is company memory, workflows, and decision making different game, bigger stakes
English
0
0
0
11
nabu
nabu@nabushhh·
@fingaz_1122 @svpino bro you’re not building software, you’re running a restaurant in your head wake up💀
English
0
0
0
17
reed
reed@fingaz_1122·
@nabushhh @svpino Have you ever ate a meal cooked by two chefs at the same time? Most of the time it’s half of what each wanted lol
English
1
0
0
39
Santiago
Santiago@svpino·
If you are a software engineer still using an IDE or a CLI to build software, you ain’t gonna make it. If you are writing prompts, or using skills, you ain’t gonna make it. Slack is the paradigm. That’s the only way forward. /s
English
83
5
331
110.7K
nabu
nabu@nabushhh·
@lalalandnfts appreciate the detailed take less explanation and fewer rules is definitely key with agents, thanks for sharing🤝
English
1
0
1
21
LALA
LALA@lalalandnfts·
What I learned is that you shouldn't explain too much to him or explain the same thing to him in different words because it confuses him, a complete and short explanation works better. Also, we shouldn't set too many rules for him and forbid him from doing many things, sometimes we have to let him use his creativity, he has a different way of thinking, he may pay attention to aspects that we didn't pay attention to and give us results that we hadn't even thought of
English
1
0
2
36
nabu
nabu@nabushhh·
they’re now training people how to work with ai properly i’ve been telling you this for months micro task for you: reply with one thing you used to feed your agent (context, files, prompts, etc) that you now realize was hurting performance best replies get will get pts let’s clean up your agents together
Anthropic@AnthropicAI

We're joining @raiseus_ai as a founding partner. RAISE US is a nonprofit coalition working to strengthen the American workforce through employer-led action, AI-enabled training, and policy innovation to support the transition to transformative AI.

English
8
4
23
607
nabu
nabu@nabushhh·
@0xkivaro exactly too much context is one of the most common mistakes good one✔️
English
0
0
0
17
KIKI
KIKI@0xkivaro·
@nabushhh giving it too much context and assuming more always meant better
English
1
0
1
41
nabu
nabu@nabushhh·
@onchain_artzy this is a great example giving too many skills and bloating context kills performance points awarded for the honest share🤝
English
0
0
1
27
Artzy
Artzy@onchain_artzy·
@nabushhh giving too many skills to claude made token usage really high and it stopped giving me the output i wanted so i had to delete all of them because i couldnt tell which skill was causing it😑
English
1
0
1
39
nabu
nabu@nabushhh·
@ivax980 this is clean af clean curated context beats dumping everything. 20 points aded🔥
English
1
0
1
23
IVA
IVA@ivax980·
@nabushhh Spot on Clean curated context beats dumping everything
English
1
0
2
56
nabu
nabu@nabushhh·
exactly. the shopping part is easy. the money movement + compliance part is where it breaks. best approach i’ve seen: let the agent propose the transaction, but every actual payment has to be confirmed on a physical ledger signer. agent never touches the keys. curious what you think about hardware in the loop for this
English
0
0
0
33
Chubby♨️
Chubby♨️@kimmonismus·
Everyone's teaching AI agents to shop. Almost nobody is solving the boring part: letting them move the money, pay across borders, keep the books, stay compliant. That's the gap Airwallex is going after, with Airi (faster agent checkout today, a real wallet in the works) and T:0 (an AI-native platform for the finance back office, still in beta). The interface gets all the attention. The money layer is what decides whether agents can do the work, and Airwallex already owns the rails for it. It’s great that the important aspects are now also being addressed with the help of AI agents.
Jack Zhang@awxjack

We’ve raised $320M at an $11B valuation, led by Addition. AI is changing how companies are built. Teams are smaller, global from day one, and using agents more. We’ve spent 10 years building the financial rails for that world. We’re now building the intelligent layer on top.

English
16
7
96
18.8K
nabu
nabu@nabushhh·
@soroshlp saw your comment on the first one and changed the whole chip to aftabe, so i think you know my thoughts already😁
English
1
0
1
33
SOROSH
SOROSH@soroshlp·
@nabushhh u think this fork gonna actually cook or just be a meme tell me your thoughts 😂
English
1
0
1
47
nabu
nabu@nabushhh·
openai shipped jalapeño iran forked it overnight introducing aftabeiño
nabu tweet media
OpenAI@OpenAI

We’ve designed and built our first AI chip: Jalapeño. Designed from the ground up by OpenAI and brought to production with @Broadcom, Jalapeño is purpose-built for the LLM workloads powering ChatGPT, Codex, the API, and future agentic products. Chips are foundational to the AI economy. Building our own expands our full-stack platform from products to models to infrastructure, and will help us scale intelligence, serve more people, and expand access to AI.

English
17
4
31
1K
nabu
nabu@nabushhh·
@JunaidAckroyd Loops are being overhyped hard. For 80% of normal workflows (summaries, drafts, quick code, data pulls) a strong single prompt or short chain is faster, cheaper, more reliable, and way less debug hell.
English
0
0
1
53
Junaid Ackroyd
Junaid Ackroyd@JunaidAckroyd·
The best engineers don't prompt anymore
English
139
11
398
63.4K
nabu
nabu@nabushhh·
@Alisina_base he had early access and using it already💀 we normal peeps have to wait
English
0
0
2
30
nabu
nabu@nabushhh·
@lalalandnfts had to update it to their latest chip💀
English
1
0
1
25
LALA
LALA@lalalandnfts·
@nabushhh Love this one more 🤣🤣🤣👌
English
1
0
1
36
nabu
nabu@nabushhh·
@navlld thanks for adding the deck🤣🤣
English
0
0
2
35
nabu
nabu@nabushhh·
@M5Stack ordered two, waiting to receive but had to oay 2.5x cause you guys wer soldout😑
English
0
0
1
116
M5Stack
M5Stack@M5Stack·
Preparing for this weekend
English
126
785
7.4K
911.4K
nabu
nabu@nabushhh·
@Alisina_base this one is the hardest challenge by far, lets see how it goes broda
English
0
0
1
25