Analyst Uttam

Data Analyst

[Analyst Uttam @analystuttam]

https://t.co/j31dgHBmBb

[Analyst Uttam @analystuttam]

Fast doesn't mean smart.
Smart doesn't mean fast.

Every model sits somewhere different on this map.

The mistake most people make:
Picking one model for everything.

[Analyst Uttam @analystuttam]

@actuallyimthe can't stop thinking about this

[Analyst Uttam @analystuttam]

Everyone thinks AI works like this:
You ask a question.
AI thinks.
AI understands.
AI answers.
That's not what happens.
Not even close.

[Analyst Uttam @analystuttam]

https://t.co/CIGy2pLib5

[Analyst Uttam @analystuttam]

Data Analyst Role is not about Dashboard only, role is moving to business strategy.

[Analyst Uttam @analystuttam]

Thumbnail Designer Job is in Danger

[Analyst Uttam @analystuttam]

OpenAI revenue: $13 billion
OpenAI operating loss: $20.9 billion
The faster they grow, the more they lose.
This isn't a company problem.
It's the economics of an entire industry where value is being created at scale and captured by almost nobody building it.
The only guaranteed winner so far: NVIDIA.
Every dollar in the system flows through their chips.
Nvidia People:

[Analyst Uttam @analystuttam]

https://t.co/ixqs4Bas1W

[Analyst Uttam @analystuttam]

"AI-powered" is the new ".com"
80% of AI wrapper startups are expected to disappear in the next 18 months.
Same story as 2001.
The bubble isn't in AI.
It's in undifferentiated products built on top of AI with no moat.
Amazon didn't die in the dot-com crash.
Google was just getting started.
The crash eliminates the copycats. It accelerates the real ones.

[Analyst Uttam @analystuttam]

$581 billion invested in AI in 2025.
OpenAI lost $20.9 billion the same year.
Both numbers are real. Both are verified.
This isn't a crash.
It's a sorting.
The infrastructure layer wins regardless of which app survives.
The wrapper layer is dying.
The value is landing with users of free products — not the companies spending billions to create it.

[Analyst Uttam @analystuttam]

$581 billion invested in AI in 2025.
OpenAI lost $20.9 billion the same year.
Both numbers are real. Both are verified.
This isn't a crash.
It's a sorting.
The infrastructure layer wins regardless of which app survives.
The wrapper layer is dying.
The value is landing with users of free products — not the companies spending billions to create it.

[Analyst Uttam @analystuttam]

https://t.co/ixqs4Bas1W

[Analyst Uttam @analystuttam]

https://t.co/ixqs4Bas1W

[Analyst Uttam @analystuttam]

🚀BIG NEWS: Claude Cowork is now on mobile & web!

Hand Claude a task at your desk → close your laptop → pick up the finished work on your phone. Claude keeps working even when your computer is off.

Key Features:
• Background & scheduled tasks (runs 24/7)
• Phone notifications for decisions
• Seamless handoff between devices
• Chat + Cowork now unified in one place

How to Access:
Beta rolling out over the next weeks, starting with Max plan users. Doubled usage limits extended through August 5.

Who Benefits:
- Busy founders & executives
- Consultants, lawyers, analysts
- Creators & researchers
- Anyone who wants AI as a real teammate

Impact:
Transforms AI from a chat tool into an autonomous coworker. More productivity, less screen time, better work-life balance.

This is how agentic AI actually changes daily work.

Who’s delegating their first big task? Drop it below 👇

#Claude #AICowork #Productivity #FutureOfWork

[Analyst Uttam @analystuttam]

This changes everything 🔥

Start a deep research + doc task on desktop, close the laptop, and actually touch grass while Claude finishes it. Wake up to a full brief + draft on my phone.

Asynchronous AI that doesn’t die when your screen does = the real agent future.

When can we throw bigger projects at it?

[Analyst Uttam @analystuttam]

65% of workers fear falling behind with AI.

Only 13% are rewarded for reinventing how they work.
That gap is the most important number in the Microsoft Work Trend Index.

The organisation's incentive structure is 12 months behind where the capability is.

Which means the analyst building the new capabilities now is doing it without metrics that recognise it, without managers who model it, without explicit permission.

That's not a reason to wait.

That's the window.

[Analyst Uttam @analystuttam]

Two analysts. Same skills. Same company.

One receives requests, builds analysis, delivers outputs.
The other asks why the request exists before starting.

One gets more directionless as the manager layer thins.
The other gets pulled into the space the manager used to occupy.

The difference isn't technical skill.

It's whether you built the judgment layer or just the execution layer.

The 56% wage premium PwC documented isn't for AI fluency.

It's for knowing what to do with it.

[Analyst Uttam @analystuttam]

https://t.co/296T4X8wpl

[Analyst Uttam @analystuttam]

Manager headcount fell 6.1% before agentic AI was even deployed.

Amazon cut 14,000 roles.
Google cut a third of managers.
AI agents in Microsoft 365 grew 15x in one year.

The middle layer isn't disappearing because AI is smart enough to manage people.

It's disappearing because AI handles the 60% of management work that was always just information passing between layers.

The question isn't whether your manager's job is changing.

It already changed.

[Analyst Uttam @analystuttam]

The smartest person in the room rarely gets the opportunity.

The most findable one does.

Build in public before you're ready.

You won't be ready.
https://t.co/BkiBTqAeir

[Analyst Uttam @analystuttam]

The people who will be fine in 5 years
aren't learning the most tools.

They're building the layer the tools can't touch.
Start here:
→ Write down everything you know
that you had to figure out the hard way
→ Name the one thing you believe
that most people in your field get wrong

→ Post one honest observation tomorrow
The first one won't be good.

Neither will the next ten.
Improvement lives on the other side of starting.