Prastuti
Pokhrel
I build data pipelines, predictive models, and full-stack tools that turn raw data into decisions people can act on. M.S. Computer Science · Python · SQL · ML · Visualization.
About Me
I'm a Data Analyst with a background in software engineering, passionate about building end-to-end data solutions from writing optimized SQL queries and cleaning messy datasets, to deploying machine learning models and crafting dashboards stakeholders actually use.
With experience across government, academia, and software, I bring a practical, impact-focused lens to every data problem. I care about making insights accessible not just technically correct, but clearly communicated.
When I'm not wrangling data, I'm building apps, contributing to open-source projects, or exploring new ML techniques.
Skills & Tools
Experience
- Queried, cleaned & validated institutional student data using SQL and Python to support decision-making.
- Produced IPEDS-sourced benchmarking reports covering degree conferral trends, employment outlook & rankings.
- Led expansion of Cardinal Clarity web app into Android/iOS, broadening student accessibility.
- Built & deployed a Django quiz platform (Cardinal Clarity) with Nginx & Gunicorn serving 100+ students.
- Designed responsive UIs with React; developed secure REST APIs; managed PostgreSQL schema updates.
- Presented platform design and educational impact at the Digital Ticket Learning Technologies Conference.
- Led 30-member team converting 500,000+ land records across 40 municipalities at 99% accuracy.
- Reduced manual processing time by 50% through automated Python workflows for image-based records.
- Coordinated SQL cross-checks and secure PostgreSQL migration for real-time centralized access.
- Built Django APIs delivering real-time health monitoring for 20+ client servers.
- Reduced API latency by 30% through query optimization and caching.
- Built LeafletJS dashboards for geolocation-based server tracking.
Projects
Logistic Regression classifier predicting university dropout risk on 4,424 students across 36 features. Production-grade sklearn Pipeline with stratified cross-validation and stakeholder-ready odds ratio insights.
Interactive dashboard analyzing 10 years of IPEDS data across enrollment, retention, completion & financial aid — benchmarking Lamar University against 5 Texas peer institutions with a radar chart, KPI cards & funnel views.
University quiz platform serving 100+ students, deployed with Django, Nginx & Gunicorn. Expanded to Android/iOS with React Native. Presented at the 2024 Learning Technologies Conference.
Led nationwide initiative digitizing 500,000+ physical land records across 40 municipalities into a centralized PostgreSQL system. Automated Python workflows eliminated manual processing and achieved 99% data accuracy.
Get in Touch
I'm open to data analyst roles, ML projects, and interesting collaborations. Whether you have a question, a project idea, or just want to connect , my inbox is open.
Send a Message →