About
I am an Applied Scientist at LinkedIn (ex-Amazon) with 8+ years of experience designing and scaling data science systems for high-stakes decisions in product, finance, growth, and operations.
Current Focus
- Interpretable forecasting pipelines for long-term impact and resource planning
- Causal inference ecosystems combining experimental and observational data
- Scalable decision systems for engagement, revenue, and operational strategy
Methods and Domains
- Causal inference: DML, heterogeneous treatment effects, synthetic control, switchback, diff-in-diff, and A/B testing
- Forecasting: strategic, operational, and causal forecasting with ML + econometrics
- Economics: digital platforms, pricing, incentives, ads, and budget allocation
Tooling
- Python, SQL, TensorFlow, scikit-learn, statsmodels
- EconML, DoWhy, Nixtla
- Flyte, SageMaker, Tableau
Profiles
- LinkedIn: https://www.linkedin.com/in/pakshingho/
- GitHub: https://github.com/pakshingho