Preslav Aleksandrov

I'm a PhD student in Computer Science at the University of Cambridge working on the fundamental architectures that will power next-generation ML systems.

What I Do

I optimize LLM throughput and investigate step-change improvements in machine learning. My research focuses on making models faster, more scalable, and practically deployable—with all claims backed by rigorous evaluation and open code.

Current focus: Architectural innovations for the next generation of foundation models.

Why It Matters

The next wave of AI won't just be about bigger models—it will be about fundamentally better architectures. I'm building the backbone technologies that make future ML systems feasible at scale.

Research Philosophy

• Fundamental: Tackle core problems, not incremental tweaks
• Scalable: Solutions that work at real-world scale
• Realistic: Rigorous evaluation, reproducible results
• Open: All research comes with working code

Let's Work Together

I'm interested in:

• Research collaborations on next-gen architectures and LLM optimization
• Industry opportunities where fundamental ML research meets production systems
• Startup funding for ventures building the infrastructure layer of future AI