Wasif Haque
Available for opportunities

Wasif Haque

Data Scientist | ML Engineer | Computer Vision Specialist

Transforming research breakthroughs into production-ready ML systems that solve real-world fraud detection challenges

4+
Years Experience
10+
ML Models Deployed
75%
Faster OCR Pipeline

About Me

What Drives Me

Data Scientist with 4+ years of experience bridging the gap between research and production systems. I specialize in developing and deploying machine learning solutions for document fraud detection, with deep expertise in computer vision, deep learning, and system architecture.

I've progressed from implementing research outcomes to independently designing ML models, leading architectural initiatives, and mentoring team members. I thrive at the intersection of research and engineering—there's something deeply satisfying about taking a promising research prototype and transforming it into a robust, performant system that delivers real value.

Whether it's shaving 75% off processing time, architecting systems for maintainability, or sitting down with a customer to solve their fraud detection challenges—I'm motivated by impact. Passionate about building performant, maintainable systems that solve real business problems.

Machine Learning

Deep Learning, CNNs, Siamese Networks, YOLO, Random Forest, PyTorch, TensorFlow

Computer Vision

OpenCV, Image Processing, OCR, Document Analysis, Feature Extraction

Software Engineering

Python, Java, System Architecture, Performance Optimization, Design Patterns

Fraud Detection

Handwriting Verification, Signature Detection, Check Analysis, Anomaly Detection

Professional Experience

Data Scientist I

ARGO Data - Fraud Detection & Analytics R&D Group Jan 2023 – Present · 2+ yrs

I develop and deploy machine learning systems for check fraud detection, working across the full ML lifecycle from research to production.

Key Contributions:

  • Built end-to-end fraud detection models including Siamese CNNs for handwriting verification, signature forgery detection using Random Forest with custom image features, and YOLO-based object detection for check field extraction
  • Led major architectural redesign transforming the image analysis application from depth-first to feature graph-based architecture, then refactored to plugin-based system—dramatically improving maintainability and extensibility
  • Achieved 75% reduction in OCR processing time through strategic pipeline optimization, making our solution competitive in the market
  • Created queue optimization tool that analyzes thousands of configurations to reduce false positives while maximizing fraud capture. This became a paid service offering generating new revenue streams
  • Delivered multiple high-stakes customer engagements, diagnosing false positive issues, presenting findings directly to clients (including Shinhan Bank), and implementing fixes with quick turnaround
  • Mentored junior data scientists and interns, helping them translate research code into production-ready systems
Python PyTorch OpenCV Tesseract Azure OCR TrOCR PaddleOCR YOLO scikit-learn

Software Engineer

ARGO Data - Fraud Detection & Analytics R&D Group Jan 2021 – Dec 2022 · 2 yrs

Served as the technical bridge between the R&D team and software development, implementing analytical enhancements into the core Java fraud detection platform.

Key Contributions:

  • Modernized model execution framework by replacing PMML with M2cgen for improved analytical model deployment
  • Optimized fraud detection performance by implementing Bloom filter solution for duplicate check detection, eliminating expensive database lookups
  • Enhanced detection algorithms including clustering improvements for transaction anomaly detection and regression tuning for serial number validation
  • Initiated ML research that led to development of handwriting style verification—my prototype's accuracy greenlit full-scale check alteration detection development
Java Spring Boot Python SQL

Graduate Teaching Assistant

The University of Texas at Dallas Aug 2017 – May 2020
  • Assisted in teaching and grading for graduate-level computer science courses
  • Held office hours to provide academic support to students
  • Courses: Advanced Algorithms Design & Analysis, Software Engineering

Interactive Demos

Experience machine learning concepts through hands-on visualization

Neural Network Visualizer

Build and visualize a neural network architecture. Watch data flow through layers in real-time.

4 neurons
Medium
Layers: 3
Total Neurons: 12
Connections: 48

Generative Art Engine

Create unique algorithmic art with mathematical patterns. Each generation is one-of-a-kind.

Medium
Elements: 0
Pattern: -
Seed: -

Skills & Technologies

Machine Learning & AI

Deep Learning PyTorch TensorFlow Keras scikit-learn CNNs Siamese Networks YOLO Random Forest Model Training & Evaluation

Computer Vision

OpenCV Image Processing Template Matching OCR (Tesseract) Azure OCR PaddleOCR TrOCR Feature Extraction Document Analysis

Software Engineering

Python Java Spring Boot System Architecture Design Patterns Performance Optimization Plugin Architecture

Data Science

Data Analysis pandas NumPy Matplotlib Jupyter SQL Customer Analytics

Education

Master of Science in Computer Science

The University of Texas at Dallas

2020

Specialization: Intelligent Systems

Focus Areas: Deep Learning, Machine Learning, Software Engineering, and Data Analytics

Bachelor of Science in Computer Science

BRAC University

2017

Graduated with Highest Distinction

Strong foundation in competitive programming and algorithmic problem-solving

Let's Connect

Location

Dallas-Fort Worth Metroplex, Texas

Send Me a Message