NFT degen turned delivery maverick. Making brand campaigns flow like butter. Building with @bricktopians
Joined December 2012
- Tweets 7,030
- Following 1,484
- Followers 2,119
- Likes 7,592
501 Photos and videos
Pinned Tweet
Jun 5
Everything changed on Tuesday.
I became a father
Mother and baby are doing well.
Watching my wife fall in love with our little boy has been the most beautiful thing I have ever seen.
Grateful for them every second ❤️
155
3
213
5,864
Jun 10
my M4 mini setup has always had the same problem:
the model can fit, but waiting token by token still makes it feel annoying
DiffusionGemma matters because it attacks the actual bottleneck.
not smaller weights.
different generation.
Jun 10
Meet DiffusionGemma!
An experimental open model that explores a fast approach to text generation, released under an Apache 2.0 license.
Moving beyond sequential, token-by-token processes to generate entire blocks of text simultaneously. Here’s what’s new with DiffusionGemma: 👇
3
91
Jun 9
No slowing down, what’s openai’s reaction will be?
Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use.
Its capabilities exceed those of any model we’ve ever made generally available.
1
34
Jun 9
>two people using the exact same claude code loop
>one reads every diff before merge
>the other approves blind and ships faster
>six months later their codebases look nothing alike
>the tool isnt the variable anymore.
>the review habit is
>nobody talks about that part.
20
Jun 8
i was today years old when i found out microwaving breast milk is a bad idea
6 days into being a dad
already getting humbled by the kitchen appliances
27
knowix.eth retweeted
Jun 5
Everything changed on Tuesday.
I became a father
Mother and baby are doing well.
Watching my wife fall in love with our little boy has been the most beautiful thing I have ever seen.
Grateful for them every second ❤️
155
3
213
5,864
Jun 1
AI PC used to mean a sticker on the box.
RTX Spark is different:
Arm CPU Blackwell GPU 128GB unified memory.
That does not prove local agents are ready.
But it finally gives them hardware worth arguing about.
3
6
106
May 30
this is the first AI PC rumour that made me stop scrolling.
> 20 CPU cores
> 6144 Blackwell CUDA cores
> 3nm
> 128GB unified memory
if NVIDIA puts that in a Windows ARM laptop, I don’t compare it to normal laptops.
I compare it to the future M5 Max for local agents.
2
5
157
May 30
AI coding is moving from editing files
to managing the whole workspace.
Threads, pinned context, isolated worktrees, parallel tasks.
The agent is becoming the project operator.
1
59
May 29
this is what finally made mobile coding click for me.
not typing code on glass.
I send 1 task from Telegram, agent works somewhere else, then I review the diff later.
Codex on Windows from ChatGPT mobile is the same loop going mainstream.
Windows users, this one’s for you.
Computer use now works on Windows, so Codex can take action on your Windows computer.
And with Windows support for Codex in the ChatGPT mobile app, you can start, review, and steer tasks on the go while work continues on your Windows machine.
An early experience, but we’re working on more ways to keep your work moving, wherever you are.
1
75
May 29
I was wrong to write off the AI PC thing completely.
most of it still feels like laptop sticker marketing.
but NVIDIA teasing “a new era of PC” before Jensen’s June 1 keynote is different.
if RTX local AI becomes a normal PC feature, the category finally has a point.
May 29
A new era of PC.
25.0528, 121.5990
1
201
May 28
Never seen a bigger clickbait than this gg
1
4
79
May 28
Ai or Aliens what will it be?
2
3
286
May 28
Time to move back to claude code?
Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Available today at the same price.
ALT Benchmark table showing how Claude Opus 4.8 compares to its predecessor and to other models on tests of coding, agentic skills, reasoning, and practical knowledge work tasks.
1
50
May 27
mcp is becoming the shared tool layer for agents.
openai is bringing remote mcp servers and connectors into the api.
hermes is bringing local and remote mcp into the agent workflow.
same direction:
agents that can actually use the tools around them.
2
1
5
136
May 27
i spent 1.5 years building software almost no one will clap for.
an internal ERP for our agency.
it runs KOL campaigns.
it keeps creator payouts sane.
what boring internal tool actually matters at your company?
2
3
37
May 27
at 1am my agent found a failing migration and gave me homework.
read the SQL diff.
check the failed test.
approve the rollback plan.
merge the PR.
I built it to take orders from me.
now I am taking orders from it.
1
2
80