I'm a Fractional CTO and hands-on software engineer with 25+ years of experience building and shipping software.

Today, I mainly help founders turn ideas and AI-built prototypes into products that are ready for the real world.

AI tools make it incredibly fast to get something working. My work starts around the point where the questions become harder:

• Is the architecture going to hold?
• Are auth, payments and customer data handled correctly?
• What happens outside the happy path?
• Can another developer understand and maintain the code?
• What will become expensive or fragile when real users arrive?
• What should be fixed now, and what can safely wait?

I'm building tools and methods to make AI-built and vibe-coded products easier to evaluate before production.

The process ranges from a local automated Launch Readiness Scan, through a deeper Trust Report, to human-led technical audits when the product needs more context and judgment.

The goal isn't to make AI development slower.

It's to keep the speed while adding the engineering discipline needed once real users, money and data are involved.

I use AI agents extensively in my own development workflow: specifications, implementation, testing, code review, documentation and analysis.

I'm particularly interested in the boundary between what agents can automate reliably and where experienced human judgment still matters.

I'm the founder of Opla, an open-source project exploring local, privacy-first AI assistants with Rust, TypeScript and Tauri.

Some projects I've built or contributed to over the years:

• Opla/opla — Local and privacy-first AI assistants using Rust, TypeScript and Tauri
• webgpu-tensors — Experimental tensor framework using WebGPU, TypeScript and Rust
• awesome-webgpu — Curated resources around the WebGPU ecosystem
• vpiper — Cross-platform real-time media processing, transcoding and streaming
• mui-color — Material UI color picker
• anzip — Async unzip library for Node.js

My GitHub also contains experiments around WebAssembly, WebGPU, video, mobile, AI and low-level systems work.

I work as a Fractional CTO / Tech Lead on products that need senior technical judgment without necessarily building a full internal CTO organisation.

Typical work includes:

• AI-built product and codebase reviews
• launch readiness and production hardening
• software architecture
• React / Next.js / Node.js platforms
• Rust, WebAssembly and performance-critical systems
• authentication, payments and data architecture
• cloud infrastructure and deployment
• AI-assisted engineering workflows
• technical leadership and team supervision

I've worked across SaaS, AI, cryptography, video, mobile, legaltech and cybersecurity.

Product & web
TypeScript · React · Next.js · Node.js · PostgreSQL

Systems
Rust · C/C++ · WebAssembly · WebGPU

AI
LLMs · RAG · AI agents · local inference · AI-assisted development

Mobile & media
Swift · Kotlin · React Native · FFmpeg · video pipelines

Infrastructure
Docker · Kubernetes · AWS · GCP · European cloud providers

I write about what happens after an AI-built prototype starts working: production risk, security, architecture, costs, privacy and the decisions that still need experienced human judgment.

You can find the articles on mikbry.com.

If the prototype works but you're not sure whether it's ready for production, start with the Launch Readiness Scan on mikbry.com.

For deeper analysis, I also offer:

• Trust Report — automated product/code review, with an optional review with me
• Audit — deeper AI-assisted and human CTO review

Website: mikbry.com
LinkedIn: Mik BRY
X / Twitter: @mikbry
Email:hello@mikbry.com

Still building. Still learning. Still shipping.