Imagine inspecting an assembly kit. The tray should contain five washers. This one contains four. Every washer looks perfect. No scratches, no contamination, no unusual shape. The kit is still wrong. That sounds obvious to a person. But a vision system trained to recognise unusual appearances may be answering a different question. Familiar doesn’t mean correct Consider a simple patch-based anomaly detector. It compares each image patch with a memory bank of normal examples, t

Why strided downsampling can make CNN predictions sensitive to a one-pixel shift—and what anti-aliasing can and cannot fix.

Why batching can change floating-point results—and how tiny score differences can become different inspection decisions.

Why equal Euclidean distances can hide different kinds of anomalies—and how covariance, whitening, and Mahalanobis distance change visual inspection.

Why non_blocking=True removes a CPU-side wait but does not guarantee GPU overlap—and how pinned memory, CUDA streams, and batching make the difference.

How dynamic resolution helps VLMs handle images from different cameras—and why patch counts, preprocessing, and compute budgets still matter.

It was a Sunday morning in Munich. I had a cup of double espresso in hand, my laptop open, and a simple question in my head: how fast can I actually finish this? Six hours later, I had my answer — and a Udacity Nanodegree certificate to prove it. The Backstory I'll be honest. I didn't set out to break any records. I'm the CTO and Co-founder of 36ZERO Vision, a Munich-based company that builds AI-powered quality inspection systems for manufacturers like Siemens, Bosch, and Mag

You're scaling, enterprise clients are asking for security questionnaires, your team doubled, someone's laptop got stolen, and you realize your "security policy" is a Google Doc from 2022 nobody reads. Sound familiar? The Right Time to Do It Don't do it pre-PMF. Do it when: You have paying enterprise customers who actually ask for it Your team is 15-30+ people and processes are breaking You're handling sensitive data (manufacturing defects, customer IP, etc.) You want to forc

In the dynamic environment of an early-stage startup, the Chief Technology Officer (CTO) must make strategic decisions to ensure rapid development, scalability, and customer satisfaction. Python, renowned for its simplicity and extensive ecosystem, is an ideal programming language for startups aiming to deliver agile, customer-focused products. This article presents a curated selection of Python packages that facilitate professional project development, emphasizing speed, rel

For decades, economic dominance was tied to technological advancements and scientific discoveries. Nations leading in innovation often led global markets. However, the future of economic supremacy hinges not only on technology but also on cultural factors. Research suggests that a nation’s culture—its openness to adopting new technologies, willingness to take risks, open-mindedness, and social trust—plays a pivotal role in determining its economic trajectory. This article exp

As a tech enthusiast and the Co-Founder and CTO of a startup revolutionizing quality inspection in manufacturing, I’ve always believed that technology’s true purpose is to amplify human potential, not diminish it. At 36ZERO Vision, we’re harnessing automation, artificial intelligence, and cutting-edge computer vision to transform how industries ensure quality—but our mission goes beyond just streamlining processes. We’re building a future where technology supports humanity, e

NDT Additive Manufacturing #36ZERO Vision