Shakleen Ishfar
- An ex-Samsung Software Engineer with 2+ years of experience.
- A Data Science Masters student at University of Rochester, graduating in December 2024.
- An active Kaggler, competing in Natural Language Processing focused data science competitions. (Got a Bronze Medal on my first competition!)
Here's a sneak peak at my GitHub activity though out the year as a heatmap
Feel free to contact me using any of my socials below! Let's dive into data patterns and shape the future together!
Work Experience
University of Rochester Medical Center
🧠 Explored the impact of pharmacological agents on drug-induced migraines by analyzing pulsatility within the glymphatic system.
- Implemented advanced image processing techniques for precise edge detection and glymph vessel area calculation.
- Applied registration methods to address data translation issues, ensuring data integrity.
- Utilized post-processing techniques, including 2D-watershed, to correct errors in glymph vessel diameter measurements.
Samsung Research and Development Institute
⌚ Designed smartwatch features to enhance end user experience.
- Mastered test-driven development by implementing 3 smartwatch features in Java and JUnit.
- Conceptualized and implemented a deep learning model in PyTorch to predict the correct audio playback mode, mono or stereo, for TWS earbuds based on positional and signal data - achieving 99% accuracy.
Intelligent Machines Limited
🏋️ Trained, tested, depolyed, and maintained machine learning models for AI-powered products.
- Constructed powerful OCR systems leveraging Tesseract and TensorFlow for document detection systems.
- Parsed financial data from bank statements using deep learning models for credit risk analysis.
🧹 Performed data cleaning, processing, and ETL workloads to supply data scientists with clean data.
- Developed image processing pipeline for boosting optical character recognition using OpenCV, Pillow, and NumPy.
- Collected social media data using Scrappy and Selenium.
Education
- Research
Experience
- Computer Science RA: Developing foundational model for DNA and genomic data.
- Neuro-science RA: Identifying distinctions in vessel pulsatility to detect drug induced migranes.
- Teaching
Experience
- Data Structures and Algorithms (Summer 2024)
- Introduction to Computational Statistics (Fall 2024)
- Course work
- Natural Language Processing
- Data Mining
- Data Science at Scale
- Statistical Machine Learning
- Extra-curricular
- Data Science Graduate Study Group Leader (Fall 2024)
- Research Experience: Abstractive text summarization using Transformers for Bangla language.
- Competitions
- 2nd among 150 teams: Intra IUT Programming Contest
- 2nd among 50 teams: National Robotech Festival Start Expo
- 29th among 200 teams: Dhaka ACM ICPC Regional Programming Contest
- Course work
- Data Structures and Algorithms
- Object Oriented Programming in C++
- Advanced Algorithms
- Pattern Recognition
- Extra-curricular
- Workshop for Game Development in Unreal Engine 4 (2018)
- Workshop for app development in Android Studio (2019)
Skills
Programming Languages
Adept in Python, SQL, and C++ programming languages, as evidenced by the collection of HackerRank badges.
Data Science Toolkit
Projects
Kaggle Automated Essay Scoring 2.0
Developed a system of ensemble regression models coupled with carefully engineered features to score essays written by students.
April 2024 - Present
Kaggle PII Detection
Developed a system using ensemble of DeBERTa models to detect personally identifiable information (PII) tokens from long text sequences. Ranked 173rd out of 2048 participants achieving bronze medal.
January 2024 - April 2024
End-to-end Tweet Sentiment Analysis
Developed an end-to-end system to ingest and process tweet data and store them in delta lake storage system. Then used MLFLow to predict sentiment using a HuggingFace BERT model.
April 2024 - May 2024
News Outlet Freedom Detection
Used LLaMa-2 to analyze sentiment and stance of news articles. Afterwards, I statistically showed significant distinctions between local and international media coverage of the same news topic.
October 2023 - January 2024
Asteroid Mining Feasibility
Utilized statistical models and hypothesis testing to analyze the feasibility of asteroid mining on NASA's JPL asteroid dataset.
November 2023 - January 2024