MARCO SAU
Working across the full AI stack — data analysis, machine learning, deep learning, and large language models, from training and fine-tuning through inference, evaluation, and deployment. Holds a Master's degree in Applied Artificial Intelligence.
Projects
Multi-agent pipelines, DeFi security, VR/AR, and LLM fine-tuning — recent things I've shipped.
View work →Videos
Technical walkthroughs from my YouTube channel, on AI/ML and the tools behind it.
Watch →Let's talk
Open to AI/ML work, collaborations, or a good technical conversation.
Get in touch →Systems administration taught me how things break. AI is teaching me how to build things that don't.
I'm a data scientist and AI systems builder working across the full stack — data analysis, machine learning, deep learning, and large language models, from training and fine-tuning through inference, evaluation, and deployment.
I currently work as an IT officer and systems administrator, managing network infrastructure and security. That hands-on systems background feeds directly into how I build and deploy AI systems in practice, not just in notebooks.
I hold a Master's degree in Applied Artificial Intelligence, with recent work spanning multi-agent LLM pipelines, DeFi security analysis, and VR/AR development.
See what I'm building →
T-Rax — Multi-Agent Unity Code Generation
A LangGraph-style multi-agent pipeline — orchestrator, planner, and executor agents — that plans and generates C# Unity scene code end-to-end.
SmartReco
A four-stage DeFi security pipeline combining Slither static analysis with REVM-based dynamic transaction replay. Tested across 17 major protocols — including Aave V3, Compound V3, Uniswap V3, Balancer V2, and Lido — with zero vulnerabilities detected and 90%+ replay accuracy.
VR Hand Interfaces on Quest 2
Replication of the CHI 2022 "Hand Interfaces" interaction technique across 11 virtual objects, migrated from the legacy Oculus Integration SDK to the Meta XR SDK, targeting Meta Quest 2.
Small-Model Fine-Tuning Ablations
Fine-tuned 1–3B parameter models (Phi-3-mini, Qwen2.5-1.5B, Gemma-2B) to study dataset-quality effects, using LoRA/PEFT and GGUF quantization across roughly 12–24 GPU hours.
Selected walkthroughs from my YouTube channel.
AI / ML
LLM training, fine-tuning, inference, evaluation & deployment · Quantization (GPTQ, AWQ, GGUF) · Distributed inference · Model serving with Ollama
Infrastructure
Docker-based ML infrastructure · Systems administration · Network security
Security
Blockchain / DeFi protocol analysis · Static & dynamic smart-contract analysis
Content
Technical tutorial production for YouTube — scripting through post-production