Resume

• Led a 5-year oscillometry–PFT ML pipeline for pulmonary patients, unifying SQL/Excel/file-based hospital data with patient matching and reproducible quality control.
• Developed OscilloFusion (multi-channel fusion with attention for oscillometry classification), reaching ~80% accuracy and delivering +5 percentage-point improvement over conventional deep-learning baselines.
• Applied robustness strategies: augmentation, synthetic generation (GAN/cGAN), and random feature representations.
• Evaluated model performance under real-world constraints including noisy labels and data drift, and implemented calibrated uncertainty estimates to support risk-aware decision-making.
• Built an interactive clinician tool to explore signals, verify predictions, inspect explanations, and swap/compare ML backbones.

• Built an ML pipeline that converts CAD drawings into structured geometric graphs for model processing.
• Designed and trained a graph neural network to recommend annotation links, achieving 94% recall and reducing reliance on expert manual annotation.
• Formulated drafting as conditional autoregressive graph inference (state-dependent sequential annotation selection).
• Integrated a human-in-the-loop workflow to replace manual entry with efficient model verification.

• Engineered a hybrid recommendation engine fusing latent matrix factorization with real-time behavioral features to personalize user feeds (PySpark ML).
• Devised spatio-temporal predictive models for delivery time estimation (ETA), integrating geographic and traffic signals to optimize logistics planning.
• Designed and analyzed A/B tests; evaluated CTR, conversion, funnel drop-off, and churn/retention cohorts.
• Developed scalable Hadoop/Spark analytics and dashboards (Tableau, Power BI) for city/vendor/item-level insights.

• Architected a multi-tenant IoT backend for high-throughput ingestion, telemetry analytics, and access control.
• Built event-driven services in Golang with Kafka for async processing, fault tolerance, and back-pressure.
• Implemented REST APIs and pipelines using MySQL/MongoDB with Redis caching; led backend + React teams.
• Implemented end-to-end fleet management pipelines: custom GPS telemetry protocols, streaming ingestion, ETA computation, and geospatial indexing.
• Led and mentored a cross-functional team of backend and frontend engineers; defined technical roadmap.

Focus: Healthcare ML, Contrastive learning, Time-series representation learning