zynQ

AI content creator | AI news, agents & tools | daily visual breakdowns

[Milo @m1lo_exe]

HOLY F*CK, Google explained why most AI agents break in production.

Annie Wang is a software engineer at Google Cloud. she sat down and built an AI agent that tests, scores and fixes itself, live, start to finish. 95 minutes, free

I watched all of it. these are the 5 moments worth your time:

12% → 11:10 - a self-improving agent is 5 boxes. 2 of them are yours
30% → 28:07 - both agents got it right. only the trajectory shows which one is right again tomorrow
55% → 52:21 - 3 ways to score an answer. only one needs no answer key
81% → 1:16:44 - the flywheel + the 3 exits you have to build
88% → 1:23:31 - 2 telemetry pipes. only one feeds the next exam

My favorite part is 28:07. two agents book the same dinner. one checks the calendar first. the other guesses the time and gets lucky. grade only the result and they look identical

The loop isn't the hard part. deciding what "good enough" means is

Learn the harness from Google, then build the full graph below:

[Milo @m1lo_exe]

THIS IS F*CKING INSANE

Someone turned Claude Code into a full quant research team. five agents instead of one chat, each with one job, and one of them exists only to kill your results

Ask a single chat if a trading rule works and it answers in 4 seconds. it never saw the chart. it never ran the test

Here’s what you’re looking at in the video:

→ left column: the whole pipeline. question → scout → backtest → verdict → sweep → skeptic → ledger → paper → runner

→ orange box: Opus 5.5 as the lead. it plans and hands work to the team

→ green box: the actual backtest. 200-day moving average rule on the S&P 500, 1999 to 2026, vs buy and hold

→ blue cards: the three attacks the skeptic runs before you ever see a result

→ bottom: everything gets written to research.md so the team never tests a dead idea twice

the team:
lead (Opus 5.5): turns a question into a plan
scout (Sonnet 5.5): checks the data before anything runs
sweeper (Sonnet 5.5): runs every variation, never just one
skeptic (Opus 5.5): tries to break every finding
librarian (Haiku 5.5): pulls strategies with evidence behind them

the numbers from the published run:
same growth. 6.8% a year vs 6.8%
worst fall -24.4% vs -56.8%
sharpe 0.61 vs 0.44
$10K → $61,526 vs $61,009
42 trades, gross of costs

Same money. less than half the crash

The strongest model goes to the skeptic, not the idea guy. ideas are cheap. killing bad ones is where the money is save and every study ends in one of three fixed verdicts: DATA VALIDATED, TRADE-OFF or NO EDGE. tested on years the rules never saw (after 2015 by default). no agent can invent a sharpe ratio it never calculated

NO EDGE still counts as a result. log it, never retest it without a new reason

this is not a money printer. no costs, no taxes, no slippage in the backtest. paper first, always

full setup with every agent file in the article below ↓

[zynQ @zynQ6969]

𝗵𝗼𝗹𝘆 𝘀𝗵*𝘁. 𝗷𝗮𝗰𝗸 𝗱𝗼𝗿𝘀𝗲𝘆'𝘀 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗷𝘂𝘀𝘁 𝗯𝘂𝗶𝗹𝘁 𝗮 𝘃𝗼𝗶𝗰𝗲 𝗔𝗜 𝘁𝗵𝗮𝘁 𝗻𝗲𝘃𝗲𝗿 𝗿𝗲𝗮𝗱 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 𝘄𝗼𝗿𝗱

no text. no ChatGPT-style brain underneath. it learned to talk the way you did, by listening to people

it's called Dessa, from Block (yes, the Cash App people)

𝘄𝗵𝘆 𝗶𝘁 𝗳𝗲𝗲𝗹𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁:
→ normal voice AI waits for you to finish, thinks, then answers
→ Dessa listens and talks at the same time
→ it says "mm-hm" while you're mid-sentence, cuts in, laughs with you
→ Block says it picks up sarcasm, hesitation and tone

𝘁𝗵𝗲 𝗽𝗮𝗿𝘁 𝘁𝗵𝗮𝘁 𝗯𝗿𝗼𝗸𝗲 𝗺𝘆 𝗯𝗿𝗮𝗶𝗻:
→ reportedly trained on ONE GB200 NVL72 rack. 72 GPUs. the exact rack from my chip teardown
→ reportedly small enough to run on a MacBook Pro with an M4 Max, faster than real time

meanwhile everyone else is burning entire data centers

𝗯𝘂𝘁 𝗯𝗲𝗳𝗼𝗿𝗲 𝘆𝗼𝘂 𝗹𝗼𝘀𝗲 𝗶𝘁:
→ it's a research preview, waitlist only
→ no benchmarks, no parameter count, no weights
→ The Verge said some demo chats felt "a little awkward and uncanny"

so the real question isn't "is it smart"
it's: would you even notice you're talking to a machine? ↓

[zynQ @zynQ6969]

@ouerbianhuaidan 100%. everyone demos the prompt, nobody demos the phone dying at 3am. what killed it most, battery or network drops?

[zynQ @zynQ6969]

𝗵𝗼𝗹𝘆 𝘀𝗵*𝘁. 𝗲𝘃𝗲𝗿𝘆𝗼𝗻𝗲’𝘀 𝗳𝗿𝗲𝗮𝗸𝗶𝗻𝗴 𝗼𝘂𝘁 𝗼𝘃𝗲𝗿 𝗵𝗼𝘄 𝗵𝘂𝗺𝗮𝗻 𝗗𝗲𝘀𝘀𝗮 𝘀𝗼𝘂𝗻𝗱𝘀

it listens while it talks. cuts in, says “mm-hm”, laughs at the right moment
no awkward pause, no robotic turn-taking
but there are 2 problems nobody’s talking about

𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝟭: 𝗶𝘁 𝗰𝗮𝗻’𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗱𝗼 𝗮𝗻𝘆𝘁𝗵𝗶𝗻𝗴 (𝘆𝗲𝘁)
→ from what Block has shown, no tool calls
→ can’t search, book, pay or open your apps
a world-class talker that can’t lift a finger

𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝟮: 𝘆𝗼𝘂 𝗰𝗮𝗻’𝘁 𝘁𝗲𝗹𝗹 𝗶𝘁’𝘀 𝗔𝗜
→ a voice on the phone used to be proof. not anymore
→ Dessa is still a waitlisted preview, but voice cloning tools are already out there
set this up with your family today:
→ pick a code word
→ hang up and call back on a number you already have
→ urgency + money = red flag
→ verify on another channel
would your parents catch it? ↓

[Kun Chen @kunchenguid]

a lot of people were impressed by dessa because it’s truly giving a next level voice conversation experience

it’s very impressive indeed - you can listen to the demos on the site and feel the magic yourself

but am i the only one seeing the massive red flags here?

1. this model can’t make tool calls.. that instantly kills almost all practical use cases, right?

2. is it net good or net bad to make AI indistinguishable from human?

https://t.co/HNKTZ78aNC

[zynQ @zynQ6969]

𝗵𝗼𝗹𝘆 𝘀𝗵*𝘁. 𝗲𝘃𝗲𝗿𝘆𝗼𝗻𝗲’𝘀 𝗳𝗿𝗲𝗮𝗸𝗶𝗻𝗴 𝗼𝘂𝘁 𝗼𝘃𝗲𝗿 𝗵𝗼𝘄 𝗵𝘂𝗺𝗮𝗻 𝗗𝗲𝘀𝘀𝗮 𝘀𝗼𝘂𝗻𝗱𝘀

it listens while it talks. cuts in, says “mm-hm”, laughs at the right moment
no awkward pause, no robotic turn-taking
but there are 2 problems nobody’s talking about

𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝟭: 𝗶𝘁 𝗰𝗮𝗻’𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗱𝗼 𝗮𝗻𝘆𝘁𝗵𝗶𝗻𝗴 (𝘆𝗲𝘁)
→ from what Block has shown, no tool calls
→ can’t search, book, pay or open your apps
a world-class talker that can’t lift a finger

𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝟮: 𝘆𝗼𝘂 𝗰𝗮𝗻’𝘁 𝘁𝗲𝗹𝗹 𝗶𝘁’𝘀 𝗔𝗜
→ a voice on the phone used to be proof. not anymore
→ Dessa is still a waitlisted preview, but voice cloning tools are already out there
set this up with your family today:
→ pick a code word
→ hang up and call back on a number you already have
→ urgency + money = red flag
→ verify on another channel
would your parents catch it? ↓

[zynQ @zynQ6969]

@martinvars The part people skip: it was reportedly trained on a single GB200 NVL72 rack. If audio-only training is that efficient, voice AI stops being a big-lab game. Small teams could build their own.

[zynQ @zynQ6969]

@akshay_pachaar Sounding human might be solved. Being useful is the open question. With no text in the loop, how does it look things up, follow instructions or call tools? Curious whether Block pairs it with a text model later or keeps it pure audio.

[zynQ @zynQ6969]

@kimmonismus The local part is the real story. A full-duplex voice model on a laptop means no round trip to a server, so the interruptions and "mm-hm"s can land in real time. Cloud voice AI has to fight network lag before it can even sound human.

[zynQ @zynQ6969]

𝗵𝗼𝗹𝘆 𝘀𝗵*𝘁. 𝗷𝗮𝗰𝗸 𝗱𝗼𝗿𝘀𝗲𝘆'𝘀 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗷𝘂𝘀𝘁 𝗯𝘂𝗶𝗹𝘁 𝗮 𝘃𝗼𝗶𝗰𝗲 𝗔𝗜 𝘁𝗵𝗮𝘁 𝗻𝗲𝘃𝗲𝗿 𝗿𝗲𝗮𝗱 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 𝘄𝗼𝗿𝗱

no text. no ChatGPT-style brain underneath. it learned to talk the way you did, by listening to people

it's called Dessa, from Block (yes, the Cash App people)

𝘄𝗵𝘆 𝗶𝘁 𝗳𝗲𝗲𝗹𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁:
→ normal voice AI waits for you to finish, thinks, then answers
→ Dessa listens and talks at the same time
→ it says "mm-hm" while you're mid-sentence, cuts in, laughs with you
→ Block says it picks up sarcasm, hesitation and tone

𝘁𝗵𝗲 𝗽𝗮𝗿𝘁 𝘁𝗵𝗮𝘁 𝗯𝗿𝗼𝗸𝗲 𝗺𝘆 𝗯𝗿𝗮𝗶𝗻:
→ reportedly trained on ONE GB200 NVL72 rack. 72 GPUs. the exact rack from my chip teardown
→ reportedly small enough to run on a MacBook Pro with an M4 Max, faster than real time

meanwhile everyone else is burning entire data centers

𝗯𝘂𝘁 𝗯𝗲𝗳𝗼𝗿𝗲 𝘆𝗼𝘂 𝗹𝗼𝘀𝗲 𝗶𝘁:
→ it's a research preview, waitlist only
→ no benchmarks, no parameter count, no weights
→ The Verge said some demo chats felt "a little awkward and uncanny"

so the real question isn't "is it smart"
it's: would you even notice you're talking to a machine? ↓

[jack @jack]

We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.

https://t.co/r0qFLdvd6c

[zynQ @zynQ6969]

@Brandel0r Full truth

[zynQ @zynQ6969]

𝗛𝗼𝗹𝘆 𝘀𝗵*𝘁, 𝟰𝟬,𝟬𝟬𝟬 𝗽𝗲𝗼𝗽𝗹𝗲 𝗷𝘂𝘀𝘁 𝘄𝗮𝘁𝗰𝗵𝗲𝗱 𝗮 "𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗯𝗼𝘁 𝘁𝘂𝗿𝗻𝗲𝗱 $𝟲𝟴 𝗶𝗻𝘁𝗼 $𝟳𝟱𝟬𝗞" 𝗽𝗼𝘀𝘁 𝗮𝗻𝗱 𝗻𝗼𝗯𝗼𝗱𝘆 𝗱𝗶𝗱 𝘁𝗵𝗲 𝗺𝗮𝘁𝗵.

So I did.

𝗧𝗵𝗲 𝗰𝗹𝗮𝗶𝗺:
→ $68 to $750,000 is 11,029x
→ That's doubling your money every single day for 13+ days
→ With zero losing days, on a bot built in 48 hours

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟭, 𝘁𝗵𝗲 "𝗮𝗿𝗯𝗶𝘁𝗿𝗮𝗴𝗲":
→ Price gaps between big crypto markets are usually a few hundredths of a percent
→ Standard exchange fees are around 0.1% to buy and 0.1% to sell
→ So most gaps are smaller than the fee you pay to take them

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟮, 𝘀𝗽𝗲𝗲𝗱:
→ Pro trading firms put their servers right next to the exchange
→ They react in microseconds
→ A bot on a laptop or iPad over home Wi-Fi shows up 100+ ms later
→ By then the gap is almost always gone

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟯, 𝘁𝗵𝗲 𝗮𝘀𝗸:
"Comment Claude, like, repost, follow, and I'll DM you the setup."
Every comment pushes the post to more people. The engagement is the product, not the bot.

𝗪𝗵𝗮𝘁 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗶𝘀 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗴𝗿𝗲𝗮𝘁 𝗳𝗼𝗿 𝗵𝗲𝗿𝗲:
→ Backtesting a strategy on real data before you risk a cent
→ Building dashboards and alerts
→ Finding out your edge disappears after fees, for free

If it printed $750K, why would anyone sell it for a comment?

[Luis Enzo @techluisenzo]

A 19-year-old Japanese student built a trading bot with Claude Code in 2 days.

Used his iPad as a second monitor.

First night: $6,732 profit.
Starting capital: $68.

Total profit so far: $750,000.

Here's how it works:

The bot scans over 50 markets simultaneously.

Syncs live BTC data from Binance every second.

Spots price errors before humans even notice.

The edge is pure speed + pattern recognition.

While traders stare at charts trying to predict the next move, his bot is already executing on mispricing across dozens of markets.

No guessing.
No emotions.
No hesitation.

Just Claude Code logic finding gaps that close in seconds.

He built the entire system in 48 hours:

→ Claude Code handles the trading logic
→ Binance API feeds real-time BTC data
→ iPad displays multi-market monitoring
→ Executes trades when arbitrage windows open

The system runs 24/7.

Every price dislocation = profit opportunity.

Most people are still trading manually, refreshing charts, second-guessing entries.

Meanwhile this 19-year-old engineering student turned $68 into $750K by letting Claude Code do what humans can't: process 50 markets instantly and execute without fear.

Why are people still trading manually?

💡I'm giving away the exact Claude Code setup for free.

24 hours only.

To get it: 👇

1️⃣ Comment "Claude"
2️⃣ Like Comment and Repost

Follow @techluisenzo

I'll DM you the complete setup.

[zynQ @zynQ6969]

𝗛𝗼𝗹𝘆 𝘀𝗵*𝘁, 𝟰𝟬,𝟬𝟬𝟬 𝗽𝗲𝗼𝗽𝗹𝗲 𝗷𝘂𝘀𝘁 𝘄𝗮𝘁𝗰𝗵𝗲𝗱 𝗮 "𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗯𝗼𝘁 𝘁𝘂𝗿𝗻𝗲𝗱 $𝟲𝟴 𝗶𝗻𝘁𝗼 $𝟳𝟱𝟬𝗞" 𝗽𝗼𝘀𝘁 𝗮𝗻𝗱 𝗻𝗼𝗯𝗼𝗱𝘆 𝗱𝗶𝗱 𝘁𝗵𝗲 𝗺𝗮𝘁𝗵.

So I did.

𝗧𝗵𝗲 𝗰𝗹𝗮𝗶𝗺:
→ $68 to $750,000 is 11,029x
→ That's doubling your money every single day for 13+ days
→ With zero losing days, on a bot built in 48 hours

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟭, 𝘁𝗵𝗲 "𝗮𝗿𝗯𝗶𝘁𝗿𝗮𝗴𝗲":
→ Price gaps between big crypto markets are usually a few hundredths of a percent
→ Standard exchange fees are around 0.1% to buy and 0.1% to sell
→ So most gaps are smaller than the fee you pay to take them

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟮, 𝘀𝗽𝗲𝗲𝗱:
→ Pro trading firms put their servers right next to the exchange
→ They react in microseconds
→ A bot on a laptop or iPad over home Wi-Fi shows up 100+ ms later
→ By then the gap is almost always gone

𝗣𝗿𝗼𝗯𝗹𝗲𝗺 𝟯, 𝘁𝗵𝗲 𝗮𝘀𝗸:
"Comment Claude, like, repost, follow, and I'll DM you the setup."
Every comment pushes the post to more people. The engagement is the product, not the bot.

𝗪𝗵𝗮𝘁 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗶𝘀 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗴𝗿𝗲𝗮𝘁 𝗳𝗼𝗿 𝗵𝗲𝗿𝗲:
→ Backtesting a strategy on real data before you risk a cent
→ Building dashboards and alerts
→ Finding out your edge disappears after fees, for free

If it printed $750K, why would anyone sell it for a comment?

[Milo @m1lo_exe]

People are decompiling 1996 games with Opus 5.5 and posting it as a win.

Almost none of them can prove the output matches the original binary.

This article fixes that. Here’s what you’re looking at in the video:

Ghidra splits the binary into header, strings and code
Sonnet 5.5 maps every function (~$0.30 a pass) and renames NOTHING
Opus 5.5 rebuilds valid_record: 4 checks, each tied to an address
tests run against the ORIGINAL binary, not against the model’s own code
4/4 match. Now it’s a spec anyone can verify.
Then the model labels a checksum loop “CRC32”. Looks right. Reads clean.
The test fails.
Seed 2166136261, multiplier 16777619. That’s FNV-1a.
Only the test caught it. The explanation never would have.

The routing that saves you money:
→ Sonnet maps ($0.30)
→ Opus interprets the hard parts ($0.60)
→ Fable 5.1 only stress-tests a theory that already passed ($1.50)

Paying Fable rates to list strings burns money. Asking Sonnet to invent a protocol from one function burns time.

And one rule in CLAUDE.md does most of the work:
OBSERVED vs INFERRED vs UNKNOWN. No address, no claim. No log, no test.

Next time a “Claude rebuilt X” post goes viral, ask for the tests.

[zynQ @zynQ6969]

𝗧𝗵𝗶𝘀 𝗰𝗵𝗶𝗽 𝗵𝗮𝘀 𝟮𝟬𝟴 𝗯𝗶𝗹𝗹𝗶𝗼𝗻 𝘁𝗿𝗮𝗻𝘀𝗶𝘀𝘁𝗼𝗿𝘀. 𝗜𝘁'𝘀 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘁𝘄𝗼 𝗰𝗵𝗶𝗽𝘀 𝗽𝗿𝗲𝘁𝗲𝗻𝗱𝗶𝗻𝗴 𝘁𝗼 𝗯𝗲 𝗼𝗻𝗲.

NVIDIA's B200 is behind a lot of the AI you use every day. I had Claude take it apart and rebuild it in code, layer by layer.

𝗪𝗵𝗮𝘁'𝘀 𝗶𝗻𝘀𝗶𝗱𝗲, 𝘁𝗼𝗽 𝘁𝗼 𝗯𝗼𝘁𝘁𝗼𝗺:
→ Cooling plate: pulls away ~1,000 W, like a microwave that never turns off
→ 2 silicon dies: 104B transistors each
→ 8 stacks of HBM3e memory: 180 GB total
→ Interposer: wires the dies to the memory
→ Substrate: fans out thousands of contacts
→ Power stages: feed the chip at under 1 volt
→ Module board: power in, NVLink out

𝗧𝗵𝗲 𝘄𝗲𝗶𝗿𝗱 𝗽𝗮𝗿𝘁:
Each die is already about as big as a chip factory can print in one go. So NVIDIA put two side by side and linked them at 10 TB/s. Software just sees one GPU.

𝗧𝗵𝗲 𝗽𝗮𝗿𝘁 𝗺𝗼𝘀𝘁 𝗽𝗲𝗼𝗽𝗹𝗲 𝗺𝗶𝘀𝘀:
The memory moves 8 TB every second. Sounds endless. But a 70B model at 8-bit weighs 70 GB, so one full read of its weights takes ~9 ms. When AI writes text, memory is the bottleneck, not math. That's why every new generation fights for bandwidth.

𝗛𝟭𝟬𝟬 → 𝗕𝟮𝟬𝟬 𝗶𝗻 𝟮 𝘆𝗲𝗮𝗿𝘀:
→ Transistors: 80B → 208B
→ Memory: 80 GB → 180 GB
→ Bandwidth: 3.35 → 8 TB/s
→ Power: 700 W → ~1,000 W

𝗡𝗼𝘄 𝘁𝗶𝗺𝗲𝘀 𝟳𝟮:
One rack holds 72 of these, wired into one giant GPU.
→ ~15 trillion transistors
→ 13.4 TB of memory
→ 130 TB/s between chips

And the next one, Rubin, is already at 336B transistors.

What should I take apart next?

[zynQ @zynQ6969]

𝗧𝗵𝗶𝘀 𝗰𝗵𝗶𝗽 𝗵𝗮𝘀 𝟮𝟬𝟴 𝗯𝗶𝗹𝗹𝗶𝗼𝗻 𝘁𝗿𝗮𝗻𝘀𝗶𝘀𝘁𝗼𝗿𝘀. 𝗜𝘁'𝘀 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘁𝘄𝗼 𝗰𝗵𝗶𝗽𝘀 𝗽𝗿𝗲𝘁𝗲𝗻𝗱𝗶𝗻𝗴 𝘁𝗼 𝗯𝗲 𝗼𝗻𝗲.

NVIDIA's B200 is behind a lot of the AI you use every day. I had Claude take it apart and rebuild it in code, layer by layer.

𝗪𝗵𝗮𝘁'𝘀 𝗶𝗻𝘀𝗶𝗱𝗲, 𝘁𝗼𝗽 𝘁𝗼 𝗯𝗼𝘁𝘁𝗼𝗺:
→ Cooling plate: pulls away ~1,000 W, like a microwave that never turns off
→ 2 silicon dies: 104B transistors each
→ 8 stacks of HBM3e memory: 180 GB total
→ Interposer: wires the dies to the memory
→ Substrate: fans out thousands of contacts
→ Power stages: feed the chip at under 1 volt
→ Module board: power in, NVLink out

𝗧𝗵𝗲 𝘄𝗲𝗶𝗿𝗱 𝗽𝗮𝗿𝘁:
Each die is already about as big as a chip factory can print in one go. So NVIDIA put two side by side and linked them at 10 TB/s. Software just sees one GPU.

𝗧𝗵𝗲 𝗽𝗮𝗿𝘁 𝗺𝗼𝘀𝘁 𝗽𝗲𝗼𝗽𝗹𝗲 𝗺𝗶𝘀𝘀:
The memory moves 8 TB every second. Sounds endless. But a 70B model at 8-bit weighs 70 GB, so one full read of its weights takes ~9 ms. When AI writes text, memory is the bottleneck, not math. That's why every new generation fights for bandwidth.

𝗛𝟭𝟬𝟬 → 𝗕𝟮𝟬𝟬 𝗶𝗻 𝟮 𝘆𝗲𝗮𝗿𝘀:
→ Transistors: 80B → 208B
→ Memory: 80 GB → 180 GB
→ Bandwidth: 3.35 → 8 TB/s
→ Power: 700 W → ~1,000 W

𝗡𝗼𝘄 𝘁𝗶𝗺𝗲𝘀 𝟳𝟮:
One rack holds 72 of these, wired into one giant GPU.
→ ~15 trillion transistors
→ 13.4 TB of memory
→ 130 TB/s between chips

And the next one, Rubin, is already at 336B transistors.

What should I take apart next?

[zynQ @zynQ6969]

@thegreatest_sv These tools cover the execution part well. The 20% they don't cover is deciding what should exist in the first place, and that's the part clients actually pay a good designer for. I think the designers who learn these repos end up shipping twice as much, not losing their jobs.

[zynQ @zynQ6969]

@0xMovez Building it part by part before flying is the right order. Seeing why a heavy-lift octo needs different motors than a 5" racer teaches more than any spec sheet. Does it show how the weight changes flight time as you add parts?

[Milo @m1lo_exe]

I asked Grok Bot how to save tokens. It lied about its own product.

Turns out the same task can cost 551,000 tokens or 12,000.

Grok Bot doesn't count your messages. It counts every time it re-reads your chat

The fixes that actually work:

> Fresh chat per task: busy bots re-read 80-90K tokens on EVERY step
> Connectors, not browser: 45x fewer tokens for the exact same job
> Slow your routines: one user went from 672 runs a week to 168. Empty runs still bill
> One task, one prompt: every "no, I meant..." makes it re-read everything
> One bot, one role: a do-everything bot drags every old topic into every new task
> Freeze the description: same prefix = cache hit, input gets 4x cheaper

Most people will just upgrade their plan. The ones who fix their setup get a month out of the same limits

Full breakdown below:

[zynQ @zynQ6969]

𝗧𝗵𝗲𝗿𝗲 𝗮𝗿𝗲 𝟮 𝘁𝗵𝗶𝗻𝗴𝘀 𝗖𝗹𝗮𝘂𝗱𝗲 𝗗𝘂𝗲𝘁 𝘄𝗼𝗻'𝘁 𝗱𝗼, 𝗲𝘃𝗲𝗻 𝗶𝗳 𝘆𝗼𝘂 𝘀𝗮𝘆 𝘆𝗲𝘀.

Trade securities. Move money between your accounts.

Everyone's talking about what Duet can do. Anthropic's full write-up has a part most people skipped: where it stops.

𝗪𝗵𝗮𝘁 𝗶𝘁 𝗱𝗼𝗲𝘀 𝗼𝗻 𝗶𝘁𝘀 𝗼𝘄𝗻:
→ Sends you a morning brief from all your connected apps
→ Checks them every few hours and messages you only when something needs you
→ Drafts replies, builds forecasts, preps payroll
→ Works in its own virtual machine with its own browser, so your laptop can be off

𝗪𝗵𝗮𝘁 𝗻𝗲𝗲𝗱𝘀 𝘆𝗼𝘂𝗿 𝘆𝗲𝘀:
→ Sending a message
→ Making a payment
→ Signing a document
→ Deleting anything
It shows you exactly what it's about to do, then waits.

𝗪𝗵𝗮𝘁 𝗶𝘁 𝗻𝗲𝘃𝗲𝗿 𝘀𝗲𝗲𝘀:
→ Your passwords or sign-in tokens
→ Your real card number. Payments go through a virtual card via Link

On top of that, a dedicated classifier checks every message so things like card numbers and government ID don't go out. And it's built to resist hidden instructions planted in emails and web pages.

Research preview, waitlist only, Pro and Max in the US. It shares your plan's usage limits, and Anthropic admits it can still misjudge what's urgent.

Would you let it handle your payments?