About Me

I'm a data reliability expert and early-stage startup operator. I've worked across the full data lifecycle, ranging from data analyst, data scientist, and data engineer, where I've seen how data fails at every stage– and then built the systems to prevent such issues.

I earned my M.S. from Stanford Medicine, where my training focused on causal inference and observational studies. I went into the program expecting to pursue medical school, and came out obsessed with data through my exposure to tech and startups in Silicon Valley. This transition was highlighted by my acceptance into Stanford GSB's Ignite program, where I learned how to go from ideation to building a VC-backed company.

Since grad school, I've worked exclusively in early-stage startups, where I analyzed sensitive health data at scale, deployed natural language processing (NLP) models into user-facing products, developed data quality algorithms that classified over a billion medical records, and built data infrastructure that supported data science teams. I've codified these skills as a LinkedIn Learning instructor, where I've taught over 40k+ students best practices on data engineering and data quality across the entire data lifecycle.

While my expertise is in data, I also realized that the limiting factor on its impact isn't usually technical: it's leadership buy-in. Thus, I took a brief detour to lead GTM as employee one at a venture-backed AI startup, Gable, and partnered with the CEO to define data contracts as a new market category and sell into my audience of ~65k+ LinkedIn followers. I spent three years on the business side: running the GTM motion that generated 10,000+ leads and 300+ meetings with senior data leaders, building champions among directors and VPs, and closing contracts that were integral to reaching our Series A. Part of this work was co-authoring the O'Reilly book "Data Contracts: Developing Production-Grade Pipelines at Scale," which garnered over 7k downloads through our lead-gen forms alone.

Today, I'm back to building via developing reliable agentic systems on top of Gable's product. I've spent my career working on every layer of this problem, and I've spent the last three years learning how to sell the fix to the people who decide whether to buy it.

Want to collaborate? I'm specifically looking to solve the problem of post-deployment AI evaluation in regulated environments. I believe this is the hardest problem in AI and has the greatest potential to positively impact society, as our critical infrastructure becomes increasingly embedded with AI capabilities.