I help teams turn data and probabilistic models into products and decisions people trust.

Most recently I was a Senior Data Scientist at Meta, leading growth analytics for WhatsApp – growth strategy, experimentation at billion-user scale, and the identity model behind how WhatsApp counts its users. As of mid-2026 I’m independent: building askdata.app and taking on select consulting engagements. Before Meta: AI product development at Strivr, product data science at Lyft, and consulting work helping organizations find the metrics that actually matter.

Things I’ve done

  • Drove WhatsApp’s growth/integrity strategy for a $400M+ registration budget. Delivered 6% of the company growth goal at roughly 1/100 the cost of other levers – outperforming higher-budget pillars and enabling WhatsApp’s first year-over-year budget reduction.
  • Scaled cross-app account linking to 300M+ accounts in six months. Led analytics for WhatsApp account linking with Facebook and Instagram across 5+ product and engineering teams, shipping 10 major cross-app features – 53% over target.
  • Designed WhatsApp’s new identity model, now the canonical way the company accounts for registering users – preserving privacy while staying consistent for returning registrations.
  • Made experimentation faster for 30+ engineers. Built AI-augmented experiment review templates with pre-pulled metrics and automated regression breakdowns, then extended those standards to other data scientists and AI analytics agents through trainings.
  • Reduced Lyft’s steady-state failed-charge losses by 62% – $24M+ in combined annual savings from an ML fraud classifier with active-learning sampling, rollout experiments, and policy tests. Promoted twice in three years.
  • Detected severe patient events 12 seconds earlier than the industry standard, at the same accuracy. Time-series work on continuous patient monitoring data for Bernoulli One.
  • Shipping with LLMs – evaluation, prompt patterns, and the long gap between demo and product. Building askdata.app now.

The arc

Data science → product → building. I’m most useful at the seam: framing the right problem, doing the analysis that settles it, and shipping the thing that moves the metric.

Let’s talk

If you’re building with data or AI, I’d love to talk – I’m currently taking on consulting engagements. hello@kpan.dev.