Shahan Ahmed / Research & Technical Writing
Field Notes
From the Writer
This blog is a working notebook for applied machine learning, OCR document intelligence, healthcare analytics, data cleaning, and research systems. The focus is practical: what worked, what failed, what the metrics showed, and what the data taught me.
Latest articleParticipation & Collaborations
Conferences, meetings, and active research work
A running record of academic programs, public presentations, and collaborative research initiatives.
Upcoming
SICSS Chicago 2026
Participant · Chicago State University · July 2026
Summer Institute in Computational Social Science focused on computational methods, networks, text, and social data.
Ongoing
Community Livelihood Risk Survey
PI / Research Lead · NGO-supported survey project
OpenDataBD
Founder / Research Data Platform · Open data initiative
Past
MSU Student Research Symposium
Poster presentation · DHS vaccination coverage analysis
Presented survey-based public health analysis using Demographic and Health Survey data.
Articles
Recent writing
Turning an Open-Access Book into a Free Audiobook with Python
How I converted Matthew Salganik's open-access Bit by Bit into a 10-hour audiobook using OCR, web scraping, and neural text-to-speech — a reusable Python pipeline for turning any open-access book into audio.
The Role of Model Selection
A detailed project write-up on cleaning noisy OCR text, detecting data leakage, comparing BERT and TF-IDF SVM models, validating on external data, and implementing Pegasos-style SVM optimization.
Why I Moved This Blog Away from MongoDB
For a personal portfolio, a static blog inside the GitHub repo is simpler, faster, and easier to deploy on Vercel.
Research Notes: Computational Social Science and Health Data
A short note on using computational tools for social research, public health analytics, and policy-relevant data work.