M. Tanzil Idrisi

CS + Math @ Beloit. Experimenting in public. Cappuccino addict.

now
  • Exploring Computer Vision, Machine learning, and Android/IOS at Beloit.
  • Working on some fun side projects and busy writing something... πŸŽ‰
about

I'm primarily a technical builder and enjoy working on creative projects. My academic background is in applied math and CS, with splashes of philosophy, physics, and psychology.

I spend a lot of time thinking about crafting tight-knit communities for builders. I'm a venture partner at Contrary and help organize Friends and Family.

In my free time, I enjoy tearing apart longform essays, practicing photography, playing ultimate frisbee, and defying gravity (by lifting heavy weights).

experience

Bursar Financial Services Beloit, WI
Data Science Intern Sep 2022 - Present
β—¦ Conducted comprehensive data analysis using statistical techniques and machine learning algorithms to identify patterns and trends in financial data and the number of students interested in coming to college.
β—¦ Created visually appealing data visualizations using Tableau and matplotlib to present complex information in a clear and concise manner.

Carnegie Mellon University School of Computer Science Pittsburgh, Pennsylvania
Machine Learning Research Interm -- Dr. Ganesh Mani August 2023 – January 2024
β—¦ Conceptualized a mobile system using Python to achieve pose-invariant 3D face reconstruction from 2D, for cancer and cosmetic cases. It creates patient phantoms, combining with internal CT scans for XR based surgical analysis
β—¦ Trained a GAN in PyTorch, using Scikit-Learn, Numpy, and OpenCV to preserve facial features in translation. Achieved 90.2% Accuracy on the LFW Dataset and fine-tuned it on CMU Multi-Pie for consistent illumination across angles.
β—¦ Used DLib extract 68 facial landmarks, mapping them onto the 3D FLAME model with SciPy, PyMesh, and Matplotlib.

KFin Technologies (RTA with AI powered Mutual Fund Analytics) Mumbai, India
Software Engineer Intern May 2023 – August 2023
β—¦ Implemented and hosted a Google Cloud OAuth 2.0 based server on AWS ECS to manage the ML lifecycle and ensure code reusability across Data Science projects. Analyzed security flaws in Basic HTTP Auth and deployed platform versions via Terraform stacks, utilizing AWS services like S3, Route 53, RDS, and Secrets Manager.
β—¦ Wrote Python scripts to track Expense Ratio data of Mutual Fund Schemes, utilizing Selenium, BeautifulSoup, Pandas, Seaborn, and AWS Athena to scrape AMFI India’s site. Implemented a custom Vector Comparison method using NLTK, spaCy, and Scikit-Learn for semantic mapping of 5M+ Historical Schemes to Current Records (80% Accuracy).
β—¦ Built and scheduled AWS Lambda Functions to validate investor data for 20+ Asset Managers, classifying anomalies and relaying them to project teams for review via email and AWS Simple Notification Service (SNS).

Franklin Templeton Investments Bengluru, India
Data Science InternFeb 2022 - July 2022
β—¦ Developed a robust web scraper to parse and extract inventory emails and load them into a database on a daily basis. Queried large-scale database using MySQL to compute key deal metrics and keep track of collateral performance.
β—¦ Extracted Price and Discount Margin from Color, and merged and updated older BWiC data with the latest numbers. Worked on clean extraction of tabular data in PDFs with the help of libraries tabula-py and Camelot in Python.

Tarsense Private Limited Remote
Full Stack DeveloperAug 2021 - Jan 2022
β—¦ Carried out the project to manage the growth of the automation platform to scale the business faster. Developed Admin Panel by using NodeJS React to improve customer support
β—¦ Worked on MongoDB database synchronization, code versioning and containerization tools such as Git, Docker, etc.
β—¦ Streamlined the API documentation process by automating Swagger UI generation from TypeScript objects and JSDoc OpenAPI specifications, resulting in a 60% reduction in documentation creation time and a 75% increase in documentation accuracy through a request validator.

thoughts
Some Things I Believe
Questions
These pages are all dynamic and update as I learn more and chat with cool folks.

updates

One of my more recent goals is to send out personal life updates every few months to those who are interested. If you're interested, fill out this form.

I also occasionaly write about topics I find interesting at Medium.


projects
Live Time Gesture Recognition: Trained high-accuracy hand gesture classifiers by transfer learning from VGG-16 and ResNet-50 CNN architectures. [demo] [github]
Exercise Tracking App: This project is Real-Time Exercise Monitoring System With Pose Detection, Mediapipe, and OpenCV. [demo] [github]
W2P: Words to Words Poem Generator: This project is End-to-End Poetry Generation from a list of topics which includes given words as their threads by using T5: Text-To-Text Transfer Transformer. [demo] [github]