I recently completed my PhD from the University of North Carolina at Chapel Hill.

My dissertation research focused on the Applications of Machine Learning in Energy and Water Utility for Climate Resilience.

I have experience with:
– Applied machine learning for energy/water/climate
– Domain adaptation research for open-source large language models
– Energy modeling/analytics
– Geospatial data science and big data
– Open data and open knowledge

Contact me: kshitizkhana48 [at] gmail [dot] com

Recent posts

• Bridging Two Worlds: Machine Learning and Operations Research in Energy Systems Optimization
  Energy systems are becoming increasingly complex, integrating diverse renewable sources while striving to meet growing global demands sustainably. Two powerful approaches have emerged to tackle these challenges: Machine Learning (ML) and Operations Research (OR). While…
• Curating LLM Tuning Data from the FineWeb Dataset for High-fidelity Domain Adaptation
  We created a post-training dataset from FineWeb dataset for high-fidelity domain adaptation of open weight LLM (Google Flan). Parameter efficient fine-tuning through prompt tuning resulted in remarkable improvement in perplexity scores as well as demonstration…
• Benchmarking power system optimization: CPU vs GPU
  Power systems are getting increasingly complicated with renewable energy integration, distributed generation, storage, increasing demand, and more. Naturally, optimizing power systems is getting more computationally intensive. While gaming and AI industries have adopted GPU to…
• Making the spectrum of ‘openness’ in AI more visible
  A (very) recent history of openness in AI Google released demos of Gemini last week with much fanfare, but no way to even test it except with a supposed integration with Bard. Mistral AI tweeted…
• Embedding Shiny App in WordPress
  I mostly code in R and Python for my data science/machine learning projects and use WordPress in my portfolio blog. In order to communicate my experiments as interactive visualizations, I can either publish those as…