Technical Skills
PyTorch
Python
Airflow
AWS SageMaker
Docker
Kubernetes
Spark
RedisPyTorchPythonAirflowAWS SageMakerDockerKubernetesSparkRedisBuilding the infrastructure that keeps ML models running in production, data pipelines, model serving, monitoring, and the glue that connects research to real users.
ML Engineering is the gap between a working notebook and a system you can rely on at 2 AM. I build that gap: automated retraining pipelines on AWS SageMaker with active learning from clinician corrections (DentalScan), LLM inference endpoints serving 6+ live hospital systems under real-time SLAs (Qure.ai), and distributed training systems achieving 3.5x speedup across 4 GPUs with PyTorch DDP. My work is production-first, monitoring, caching, fallbacks, and deployment are part of the design from day one.
PyTorchPythonAirflowAWS SageMakerDockerKubernetesSparkRedisPyTorchPythonAirflowAWS SageMakerDockerKubernetesSparkRedisSorted by most recently updated on GitHub
Loading projects...
AI Solutions Engineer Intern
Mar 2026 - May 2026
Operated and maintained clinical AI inference endpoints supporting 6+ live EPIC/FHIR-integrated hospital systems under real-time US time zone SLAs. Managed model deployment lifecycle including testing, staged rollout, and incident response for radiology AI and EMR extraction systems.
Machine Learning Engineer Intern
Dec 2025 - Feb 2026
Engineered automated retraining pipeline on AWS SageMaker incorporating dentist-corrected active learning labels across 50K+ intra-oral images. Built 6-category CV classification system (gingivitis staging, plaque detection, recession classification) with continuous model improvement in production.
Graduate Research Assistant
Jun 2025 - Present
Built end-to-end ML pipelines for 6G wireless path planning: data generation (Sionna simulator), feature engineering, LLM fine-tuning, and evaluation. Managed 10K+ supervised training examples from simulation to model training to inference benchmarking.
Amazon Web Services • Dec 2024 - Dec 2027
Microsoft • Aug 2025
From SageMaker pipelines to live inference endpoints. Let's talk.