Khan Academy Externship via Extern
Early-career creator focused on learning, hands-on projects, and growing technical skills.
A collection of early projects, learning exercises, and experiments that show my skills, interests, and progress as I build experience.
Consolidated and analyzed 21,724 news articles; built reproducible preprocessing and modeling workflows and evaluated logistic-regression and hybrid models, achieving validation AUC up to 0.862. Diagnosed potential data leakage in a transformer-based modeling workflow, demonstrating model-risk aware
Built and evaluated a semantic recommendation system on GoodBooks-10k using sentence embeddings; achieved Precision@10 of 0.820 versus 0.480 for a TF-IDF baseline and NDCG@10 of 0.817 versus 0.383. Compared competing approaches using multiple performance metrics, balancing predictive relevance, dive
Early-career creator focused on learning, hands-on projects, and growing technical skills.
I am Sanjay Kshetri. I am at the start of my career and exploring opportunities to build practical experience, learn new skills, and contribute to real projects.
Externships
Khan Academy Strategic Partnerships & Growth Externship
In progress
Experience
Mathematics Educator / Data-Informed Instruction
Harford County Public Schools - Edgewood High School · Feb 2025-Present
Applied Data Science Fellow
Springboard Data Science Career Track · 2025-2026
Mathematics, Statistics & Physics Educator
Threefold Educational Foundation / Green Meadow Waldorf School · Jan 2022-Jan 2025
Graduate Research Assistant
Montclair State University · Aug 2020-May 2022
Education
Montclair State University
M.A., Psychological Sciences · Class of 2025
Montclair State University
Graduate Certificate, Advanced Quantitative Methods · Class of 2025
Ramapo College of New Jersey
B.A., Psychology · Class of 2019
Tribhuvan University
B.A., Mathematics & English Literature · Class of 2016
Skills
Consolidated and analyzed 21,724 news articles; built reproducible preprocessing and modeling workflows and evaluated logistic-regression and hybrid models, achieving validation AUC up to 0.862. Diagnosed potential data leakage in a transformer-based modeling workflow, demonstrating model-risk aware
Built and evaluated a semantic recommendation system on GoodBooks-10k using sentence embeddings; achieved Precision@10 of 0.820 versus 0.480 for a TF-IDF baseline and NDCG@10 of 0.817 versus 0.383. Compared competing approaches using multiple performance metrics, balancing predictive relevance, dive
Designed and analyzed a quantitative study of 194 U.S. adults; constructed measures and applied regression, path modeling, mediation, and bootstrap inference to test competing behavioral hypotheses. Prepared validated analytic data and translated multivariate results into a thesis-length research re
✅ Verified by Extern · ⏱️ In progress
100M people a year learn for free on Khan Academy. Someone pays for that. Use AI to source and pitch the deals that fund it next.
The project sourced and evaluated prospective funding partners using AI-driven research and presented the findings as concise outreach pitches. The work combined automated partner discovery with manual vetting and crafted pitch language for selected prospects.
I used AI tools to discover candidate partners, reviewed each prospect manually, and wrote concise outreach pitches based on that combined research.
