The Architect of Constraints
Machine Learning Engineer & Quantitative Researcher.
Building systems that thrive where data is sparse and compute is expensive.
Default to Constraints.
"I learned to ship ML with no paid data and minimal compute. No assumption is sacred unless the data can survive without it.
The best architectures are honest about what they don't know, and they fail in visible, reversible ways."
The Human
- Running To think.
- Cooking To slow down.
- TV To escape.
- Sociology To understand failure.
PROJECT GENESIS
From "Tango" to "Scarcity": An Architecture of Necessity.
The Data Void
Failed to build a Macro-Policy Sim because no data existed.
→ Solution: Online Learning.
The Privacy Wall
Needed sensitive user data without seeing it.
→ Solution: Federated Learning.
The Cold Start
System started with zero context.
→ Solution: Meta-Learning.
The Dimensional Chaos
Unpredictable data shapes.
→ Solution: Mixture of Experts (MoE).
Tech Twirls
Instincts over Syntax
"Mentoring women in tech to build reliable engineering instincts. We are building a community that survives the hype cycle."
Interested in collaborating? Use the site form.