VIKRAM DIXIT K

I am an economist and engineer building the next generation of economic measurement systems. Currently an Independent Researcher, I was previously a Postdoctoral Research Fellow at Stanford University’s Golub Capital Social Impact Lab and a Senior Software Engineer at Hortonworks (Cloudera).

I possess a rare combination of rigorous theoretical training (PhD, Boston University) and production-grade engineering skills. My research combines frontier methods in causal inference (Shift-Share IVs, RCTs) with Natural Language Processing (LLMs, BERT) to quantify the diffusion and impact of technology on the economy.

Current Research: My latest work investigates the "Realism Gap"—the divergence between AI’s technical capability and its economic adoption. I constructed the Verified AI Trust Friction Index by deploying a 32-billion parameter LLM pipeline to process thousands of corporate 10-K filings. This work quantifies how algorithmic liability risks act as a binding constraint on automation.

Previously, my research on Firm Risk Networks used textual analysis to isolate shared vs. unshared risk factors across 30,000+ public firms, demonstrating that text-based risk measures can significantly improve asset pricing models.

Engineering & Approach: Unlike traditional economists, I build my own data infrastructure. With 7+ years of experience as a Senior Software Engineer architecting petabyte-scale systems, I am comfortable working at the intersection of structural estimation and big data engineering. I don't just analyze datasets; I build the pipelines to create them.

Personal: Outside of research, I am a builder and a tinkerer. I recently developed a voice-based journaling Android application—loosely based on the Jetsons’ robot diary—that uses on-device AI to track physical and mental health trends. When I'm not coding, you can find me maintaining my racing bike, reading everything from Calvin and Hobbes to algorithmic trading, or geeking out on new tech.