Skip to content

About

I'm Dr. Saad Laouadi, and I build AI systems that actually work in production.

Most ML projects never make it past notebooks. I specialize in the hard part—taking models from research to production and making sure they stay there.

I also teach programming, machine learning, deep learning, MlOPS and other data science related topics.

What I Do

I design and deploy production ML systems across multiple industries. My work spans agriculture, healthcare, and cybersecurity, but the underlying challenge is always the same: building systems that are reliable, scalable, and maintainable.

I work with Python, Rust, Java, JavaScript, TypeScript, and R—not because I'm collecting languages, but because production systems need the right tool for each job. Python for ML prototypes, Rust for performance-critical code, TypeScript for robust APIs. You get the idea.

Background

I hold a PhD in Applied Economics with a focus on quantitative methods and econometrics. My academic background in statistical modeling turned out to be surprisingly useful for machine learning—turns out the math is the same, just the applications changed.

Over the past seven years, I've worked on everything from agricultural pest detection systems (preventing millions in crop losses) to healthcare analytics platforms to real-time cybersecurity threat detection. The domains are different, but the engineering principles stay constant.

Current Work

Right now, I'm consulting independently on ML system architecture and production deployment. I also teach—explaining complex AI concepts in ways that actually make sense to both technical teams and business stakeholders.

Recent projects include building a full-stack agricultural intelligence platform that serves 500+ users, developing text processing systems handling 50K+ documents monthly, and creating production ML pipelines that run reliably at scale.

This Blog

I started this blog to share what I've learned building production systems. You'll find practical insights on:

  • Taking ML from notebooks to production
  • Multi-language development strategies
  • MLOps and deployment best practices
  • Real lessons from shipping AI systems

No theory for theory's sake. Just what actually works when your code needs to run 24/7.

Let's Connect

If you're working on production ML systems or thinking about how to get there, reach out. I'm always interested in talking with people building real systems.

The Polyglot Engineer - because production ML is 90% engineering, 10% ML.