Yogananda B

Building systems for a world where software works alongside humans.
AI • Enterprise Technology • Automation
Thinking long-term.

[Yogananda B @byogananda6346]

@Dheemanthredy Great @Dheemanthredy

[Yogananda B @byogananda6346]

@CompleteSkeptic The real risk isn't AI replacing developers, but building products that a model update can make obsolete. Own your logic, retain control, and use AI to amplify human creativity, not make your work disposable.

[Yogananda B @byogananda6346]

@AnthropicAI When writing code becomes cheaper than coordinating it, the rules of software engineering change.

If AI can build at scale, what becomes the real advantage: writing code or knowing what’s worth building? 👀

[Yogananda B @byogananda6346]

@oli_so_no Hey @oli_so_no ! I'm an engineer exploring AI and building things around it. Always happy to connect with people building cool stuff! 🤝

[Yogananda B @byogananda6346]

@ayesha_fatiima Not a dumb question at all! They make money through Telegram Premium subscriptions, privacy-friendly ads in big public channels, and selling premium usernames. A small percentage of paying power users basically covers the storage costs for all the free users

[Yogananda B @byogananda6346]

@hridduSP X=50
Sum of all sides =360
X+90+60+(180-20) =360
X=50

[Yogananda B @byogananda6346]

@Saanvi_dhillon A 10GB file shrinks to 2GB. The other 8GB didn’t vanish. It was redundancy: repeated bytes and predictable patterns written the long way. Lossless compression just stops paying full price for that. Unzip it and the 10GB comes back.

[Yogananda B @byogananda6346]

@vibeonX69 Macs with 16GB often feel faster than Windows laptops with 32GB because of unified memory (CPU + GPU share the same ultra-fast pool), no data copying, smarter macOS compression/swapping, and tight chip + OS integration. Raw GB numbers don’t tell the full story.

[Yogananda B @byogananda6346]

@jiya_3063 WhatsApp is free for users on purpose.
It makes money from businesses:

• Pay-per-message (orders, promos, support)
• Click-to-WhatsApp ads on IG/FB
• Ads in Updates tab + payment fees in some countries

Private chats stay ad-free & encrypted.

[Yogananda B @byogananda6346]

I'm still early in this journey.

But the direction I'm exploring is becoming clearer:

LLMs → Agents → Edge AI → Physical AI

I'll be documenting the experiments, failures, things I learn, and the questions I can't answer yet.

Let's see where it leads.

[Yogananda B @byogananda6346]

My AI journey moved from LLMs & RAG → AI agents → Edge AI.

Now I’m exploring what happens when models move beyond APIs and into the real world — running locally on Jetson.

Models → Agents → Edge AI → Physical AI

@grok, where does this journey go next?

[Yogananda B @byogananda6346]

And this is where my interest is heading now:

Physical AI.

Models that don't just generate an answer, but can perceive, reason and eventually act in the physical world.

That's a very different problem from simply calling an API.

[Yogananda B @byogananda6346]

And this is where my interest is heading now:

Physical AI.

Models that don't just generate an answer, but can perceive, reason and eventually act in the physical world.

That's a very different problem from simply calling an API.

[Yogananda B @byogananda6346]

That brought me to Edge AI.

Running models locally. Working with open-source models. Understanding quantization, inference, memory constraints, and latency.

And that's where things got much more interesting.

[Yogananda B @byogananda6346]

@GithubProjects https://t.co/RWmQN9D0Rc

[Yogananda B @byogananda6346]

I just published Running AI Models Locally: What I Learned After Moving Beyond the Cloud https://t.co/kSDKTJOyDK

[Yogananda B @byogananda6346]

Check out my latest article: Graph Engineering: Where AI Agent Development Is Heading https://t.co/bOppuJkq8t via @LinkedIn