I am a Data-driven engineer who wrangles raw, high-volume data into scalable ML models and real-time dashboards, helping in cutting busy-work, uncovering insights, and enabling teams to act on facts instead of hunches.
For a tailored résumé specific to ML or DE role, please feel free to reach out via email.
I’m a Computer Science grad student at Arizona State University and a hands‑on data & ML engineer. Recent stints include building cloud‑scale data pipelines at Alcon, training vision models in a deep‑learning lab, and leading the telemetry team for a eATV. I love turning messy data into clean insights and give out real‑world impact using various ML models.
Developed a full-stack Flask application and orchestrated ETL pipelines on AWS, processing over 2TB of data, increasing operational visibility by 40% and reducing project development cycle time by 43%.
Prototyped a CNN-based population density model using 9GB of satellite imagery on AWS SageMaker, improving urban planning resource allocation accuracy by 45%.
Design, test, and integrate the electrical powertrain into an ATV to compete at BAJA SAE, landing among the top 20 teams.
A Python package simplifying direct data ingestion into Apache Iceberg tables using PySpark. It streamlines schema creation, data deduplication, snapshot management, and optimization, leveraging Iceberg's capabilities for high data quality and performance.
Designed a personalized video suggestion system using deep feature extraction and clustering, achieving 97.2% similarity search accuracy and improving execution speed by 18% through a streamlined pipeline.
.png)
Engineered an AI border-security platform analyzing four 1080p video streams with YOLOv8, achieving 94% recall, reducing manual monitoring by 80%, and issuing real-time alerts within 1.3 seconds, earning Runner-up at the ASU Voxel51 Visual AI Hackathon.
Developed a regression-based ML model with automated ZenML/MLflow pipelines, achieving 92% accuracy, reducing deployment time by 60%, and delivering scalable, production-ready code for cloud deployment.
DooFy enhances CAPTCHA technology by integrating image and drawing-based verification, achieving a 95% bot detection rate while improving user accessibility by 30%.