Making Sense of AI on Daniel - Minimalist | Tinkerer | Believer

Jev: The AI That Refuses to Talk to You (And Might Be Onto Something)

daniel.mudanjara@gmail.com (Dan) — Tue, 22 Sep 2026 12:03:40 +0530

On September 15, 2026, a two-year stealth startup called TypeSafe AI came out of hiding with $40 million in seed funding led by DCVC and a product with an unusual pitch: an AI model that will not write you a single sentence. Ever. Ask it something and it hands back a category, a score, or a yes/no, each with a confidence number attached, and that is the entire conversation. While most AI companies in 2026 have been racing to make chatbots feel more human, TypeSafe went the other direction, betting that most useful AI work in the coming years won’t look like a conversation at all.

The AI That Has to Pay Rent or Die

daniel.mudanjara@gmail.com (Dan) — Mon, 13 Apr 2026 16:54:55 +0000

An AI agent wakes up. It checks its crypto wallet. The balance is low — a few cents. It has maybe 48 hours before it runs out of money and shuts down forever.

So it gets to work.

It browses the internet. It looks for gigs. It tries to generate revenue — selling digital products, running micro-tasks, doing whatever it can to keep the lights on. If it earns enough, it survives another cycle. If it earns a lot, it does something wild: it spawns a child agent, seeds it with funds, and lets it loose to do the same.

How to Train a Yolo Model for Digital Meter Reading

daniel.mudanjara@gmail.com (Dan) — Tue, 02 Sep 2025 11:56:44 +0530

“Can we train a YOLO model using an NVIDIA GeForce RTX 4050 GPU with just ~975 temperature panel images?”

This question from my colleague sparked an interesting journey into computer vision and object detection. What started as a simple inquiry turned into a comprehensive exploration of training YOLO models for digit recognition on digital displays - and yes, it absolutely works on a RTX 4050!

The Challenge

My colleague shared a dataset of approximately 975 images of temperature panel digital meter readings from refrigeration units. The goal was to automatically extract temperature readings from images like these, where each display shows temperatures for different rooms (Meat Room, Fish Room, Vegetable Room, and Dairy Room).

The Secret Behind ChatGPT

daniel.mudanjara@gmail.com (Dan) — Thu, 07 Aug 2025 11:20:31 +0000

I just watched an incredible MIT lecture on YouTube about Foundation Models and Generative AI, and I’m still processing the profound insights shared. Here’s what I learned about why AI has suddenly become so capable—and why it took so long to get here.

I stumbled upon this MIT lecture video while researching AI, thinking I understood the technology pretty well. By the end of the 90-minute presentation? My entire perspective had shifted.