About

I'm a software engineer, mostly on the backend — services, proxy infrastructure, and traffic-routing systems that had to keep working under unreliable upstream conditions. The kind of work where the interesting problems are latency budgets, failure modes, and the part of the system you assumed was fine.

The pivot toward AI / ML happened in the honest direction: I kept building RAG and inference systems at work and noticed that the people shipping them well were backend engineers who had learned enough ML to be dangerous, not ML researchers who had picked up systems. The shortage in this industry isn't people who can fine-tune a model; it's people who can put one in production and keep it there. That felt like a place I could be useful.

Right now I'm going deeper on three fronts:

• Retrieval — building aegis-rag as scaffolding to actually benchmark retrieval strategies against each other, rather than shipping whatever came out of a tutorial.
• Local inference — running quantized open-weight models on consumer hardware and writing down what works. Eventually this becomes the generation backend for the RAG work.
• Computer vision — training YOLO on custom datasets, mostly to learn what actually determines mAP in practice (it's the dataset, not the architecture).

I don't call myself an AI researcher or an ML scientist. I'm a backend engineer who builds AI systems and is honest about where I am with the ML side. Both halves are load-bearing — backend experience is a real asset for this kind of work, and I'd rather lead with both than hide one.

What I'm shipping

I run Argus Intelligence — a small AI-tooling brand for the EU compliance space. The first product is the EU AI Footprint Scanner, an AST-based static analyser that flags AI/ML library use across a Python codebase, classified into simplified EU AI Act risk tiers. It's aimed at EU SMEs facing the August 2026 GPAI deadlines without enterprise-grade GRC tooling.

The Argus name comes from the hundred-eyed Greek guardian — same vigilance metaphor as a continuous PR scanner.

Background

Career started in backend engineering: services in Java, Python, and Node.js, with time spent on PHP-to-JVM migrations and on event-driven infrastructure. Biggest long-running project has been the edge routing proxy — an edge routing layer built on OpenResty with a Couchbase routing store, designed for high-frequency endpoint rotation and session-aware proxying across fast-changing upstream state.

What's next

Post the local-model benchmarking numbers. Ship a real RAGAS evaluation run on aegis-rag. Write up the ablation results from the YOLO work once the dataset is stable. Keep the /now page honest about where the work actually is.

Contact

Email: hi@zhen.ee
LinkedIn: linkedin.com/in/zhenee