Vaibhav Bhandari
I build and study intelligent systems that have to work in the real world.
Founding engineer, Mapsense → Apple · ~10 years of ML and data systems at Apple Maps · now independent, in Indore
Today I'm investigating how probabilistic AI becomes dependable under real-world constraints — latency, privacy, limited compute, offline operation, and the physical world.
Now
- Independent Research — First write-up published. No experiments run yet.
- Knos Digest — Shipping daily; has run for months with real users.
- Latest note: Small AI, Real Systems — published, with the reasoning it came from.
Latest work
Dependable intelligence in real systems — written up as it happens, including what doesn't work.
- 2026-08-18Small AI, Real Systemsworking notes
What a week in Bangalore changed in my thinking — why the questions enterprise builders ask about small AI are the same ones that shaped an agentic platform at Apple.
Also running
Long-running systems and personal instruments. Subordinate to the research on purpose — they're here because they still run, not because they're the point.
Knos Digest
A personalized, high-signal daily digest of tech news.
Anthropic's best AI model struggles to attract users as cheaper tools thrive
7 stories · AI Research
State Overflow
A data scientist's lens into everyday decisions.
The $80 Hike In Vietnam That Changed How I Think About Vacations
Understanding tourist traps through statistics
May 2026Metrics Dashboard
Observing daily habits without a productivity app.
Last week
Routine score = 0–4 based on daily habits (family, fitness, sleep, work).
Reading
Reading is an input to the research, not a shelf. A few books I keep returning to:
Steve Jobs
Sapiens
The Fountainhead
The Sirens of Titan
The Mind Illuminated