Software Engineer · AI / Embedded Systems · Stuttgart, Germany
MoMORSY
cerebrovault.dev — field notes on building AI that ships
Connect
LinkedIn GitHub Medium Coursera
Featured Project
100 Days of AI

A complete path from zero to hireable — 8 training modules covering fundamentals through advanced AI, followed by 14 market-validated specializations. Free resources only.

8
Modules
60
Training Days
14
Specializations
100
Days Total
PythonPyTorchLLMsMLOpsFree Resources
View Learning Path →
Interests
Edge AI
Autonomous Systems
ADAS Perception
LLM Engineering
On-device ML
Embedded Systems
Open Source
Projects
Cerebrofy
Give your AI agent a brain, not a firehose.

AI-powered codebase intelligence CLI — compresses repo context into structured summaries so agents understand large codebases without drowning in tokens.

pip install cerebrofy
v1.0.5 Python
Wadjet-Link
Automotive Ethernet validation framework.

Open-source framework for live packet capture, protocol decoding, and automated testing on Automotive Ethernet networks. Built for embedded systems validation in automotive-grade environments.

git clone github.com/mm0rsy/Wadjet-Link
C++
Shura
Multi-repo AI orchestration with autonomous teams.

Multi-agent, multi-repo orchestration for Claude Code. Each repo gets its own autonomous team — Engineering Manager, Product Owner, and Developer — coordinated by a Program Manager agent.

git clone github.com/mm0rsy/shura
JavaScript
GitHub Activity559 contributions this year
From the Blog
Latest Writing
All posts →
19 Jun 2026
Why Asynchronous Chat Breaks Production Debugging for Complex Embedded Problems

Asynchronous text communication, though ubiquitous, introduces a critical flaw in production-critical software development: cognitive latency. Defaulting to high-volume, low-fidelity chat for complex technical decisions in embedded systems or AI pipelines ironically leads to significant delays an...

embedded systemsai pipelinescommunication
20 May 2026
The Implicit State Tax: Why Variable Inheritance Chokes GitLab CI Scalability

Global variable inheritance in GitLab CI pipelines often feels like a shortcut. It is not. This implicit state tax demands excessive cognitive load, creating debugging nightmares, and pipeline performance bottlenecks, especially in microservice architectures.

GitLab CICI/CDDevOps
20 May 2026
The Token-Flush Latency: Why 'Full-Duplex' Voice AI Needs More Than Concurrent Streams

Full-duplex voice AI promises seamless interaction, but building production systems with sub-100ms latency demands more than just overlapping ASR/TTS. The core challenge lies in managing token-flush latency during human interruptions, where an LLM must instantly adapt its generation rather than s...

AIEmbedded SystemsLow Latency