About
Hey, I’m Yudhiesh 👋
I’m an ML engineer who believes that machine learning should actually work in production, not just in Jupyter notebooks.
Currently, I’m architecting ML infrastructure for quant trading systems at Trontal Group, where milliseconds matter and reliability isn’t optional. Before this, I built ML platforms at MoneyLion’s Machine Learning Labs, helping dozens of data scientists ship models that serve millions of users.
What I Do
🚀 Production ML at Scale
From serving LLMs handling thousands of requests per second to building MLOps platforms that don’t break at 3 AM. I specialize in the practical work of making ML systems reliable, scalable, and actually useful.
📝 Writing About Real ML
This blog is my engineering notebook made public. No “10x AI productivity” hype, no “AGI is coming” speculation. Just tested lessons from the trenches of production ML. Expect deep dives into:
- Vector search optimization (and why filters are its Achilles heel)
- Feature flag strategies for ML deployments
- Synthetic data generation for LLM safeguards
- The practical reality of ML infrastructure
🎓 Teaching What Works
I’ve taught 1500+ engineers worldwide through hands-on courses that focus on what actually matters in production:
- LLMs in Production - Beyond prompt engineering to real deployment
- MLOps: From Models to Production - The full lifecycle, including the challenges
- DevOps Crash Course - Because ML engineers need ops skills
My Philosophy
Good ML engineering is invisible. Users don’t care about your model architecture. They care that predictions are fast, accurate, and available. That’s what I optimize for.
Failures teach more than successes. Every post includes what went wrong, not just what worked. Because that’s how we all actually learn.
Code > Slides. Everything I write includes working examples. Theory is nice, but git clone
and docker run
are better.
Let’s Connect
I’m always interested in discussing production ML challenges, especially around:
- Scaling vector databases
- LLM deployment strategies
- ML platform architecture
- Making data scientists more productive
Find me on:
- 💻 GitHub - Where the code lives
- 💼 LinkedIn - For professional connections
- 📧 Email - For interesting opportunities
Currently in Kuala Lumpur, Malaysia 🇲🇾