Tempe, Arizona, United States
I am a passionate Ph.D. researcher in Computer Science at Arizona State University’s School of Computing and Augmented Intelligence (SCAI), with 7+ years of experience building human-centered AI systems. My research intersects computer vision and human-computer interaction (HCI), focusing on wearable, sensor-driven solutions for assisted living, sports, and healthcare. ► 7+ years in multimodal sensing, human activity recognition, and real-world HCI applications. ► Collaborated with FedEx, Pizza Hut, Adidas, Amazon, and Edgenuity on industry challenges in data science, AR, and HCI. ► Awarded WearTech and Global Sport Institute grants ($40k total) as Co-PI for wearable tech projects. ► Recognized as an NSF Research Trainee for Citizen-Centered Smart Cities and Smart Living. ► Co-authored a patent and first-authored peer-reviewed papers at top HCI and computer vision conferences. ► Reviewed 120+ papers and served on program committees for leading conferences (CHI, HRI, ETRA). Driven to build intuitive, impactful, human-centered technologies, I welcome opportunities to collaborate on innovative AI-powered systems. Career Highlights ★ Developed wrist-worn camera system for real-time human activity recognition, reducing medication adherence errors. ★ Co-led multimodal sensing solutions for AR in sports with FedEx. ★ Led neuromorphic anomaly detection research for Mercedes-Benz Vision EQXX Concept, achieving 4x energy efficiency. ★ Built and deployed scalable AI models on AWS and GCP for health and wellness. ★ Mentored 45+ students in applied AI and HCI research at ASU’s CUbiC, iLUX, and ANGLE labs. Areas of Expertise ▸ Data Science & Machine Learning Applications ▸ Computer Vision & Multimodal Sensing ▸ Human-Computer Interaction (HCI) ▸ Wearable Technologies & Affective Computing ▸ Augmented Reality (AR) & UX/UI Design ▸ Project Management & Research Mentorship Technical Skills ▸ Languages: Python, SQL, Java, JavaScript, HTML/CSS, Bash ▸ Frameworks & Libraries: TensorFlow, PyTorch, scikit-learn, Spark, Hugging Face Transformers, LangChain, OpenAI API, Claude (Anthropic), LLaMA, Mistral ▸ Cloud Platforms: AWS (EC2, S3, SageMaker, Bedrock, Lambda), Google Cloud (Vertex AI), Azure OpenAI ▸ Tools & DevOps: Docker, Git, HPC, Jupyter, Tableau, Elasticsearch, Weights & Biases, MLflow, Kubernetes, Streamlit, Gradio, FastAPI, Flask ▸ Vector DBs & RAG: FAISS, Pinecone, Weaviate ▸ Specialized Sensors: Eye Tracking, GSR, EEG, ECG, FSR ▸ AR/VR Development: Microsoft HoloLens, OpenCV, ROS ▸ Other: RLHF, LoRA, QLoRA, Prompt Engineering, MLOps, Ethical AI Considerations
• Developing advanced machine learning and image learning algorithms for PERACTIV, a novel wrist-worn camera device. • Collaborating as a Co-PI for the Global Sport Institute Grant to understand and develop valuable tools for sports enthusiasts and old age adults living alone track to physical activities using the novel wrist-worn camera device. • Collaborating as a Co-PI for WearTech to develop useful pill-taking tools for people with old age living alone using the novel wrist-worn camera device.
FSE 100 - Introduction to Engineering (Fall 2020, Fall 2022, Fall 2023, Fall 2025) • Taught a total of 120 students engineering concepts. • Held in-class labs and lectures. • Graded homework assignments and in-class participation activities.
CSE 463: Intro Human-Computer Interaction (Spring 2019, Fall 2019, Spring 2020, Spring 2023) • Taught over 1300 students Human-Computer Interaction (CSE 463) concepts like prototyping, usability principles, and heuristics. • Assisted in grading homework assignments, exams, and in-class participation activities. • Assisted the professor in the course structure.
• Validated the Hiji Band wearable system for real-time fall detection and anomaly recognition by training 12 Conv1D autoencoder models across 4 sampling frequencies (100Hz–10Hz) and 3 step sizes, improving robustness under varying deployment conditions and real-world signal degradation. • Tested Hiji Band across 3 body placements (arm, waist, shin) using over millions of time-series samples, analyzing reconstruction error trends to assess sensitivity to body-worn position and optimize sensor placement strategies. • Developed a scalable evaluation pipeline supporting multi-rate sequence generation, unsupervised anomaly detection, and 99th percentile-based thresholding, with results visualized through custom plots highlighting acceleration anomalies aligned with timestamped ground truth events.
• Designed and implemented AI-powered documentation for Boomi Agentic Scribe, reducing manual documentation efforts by 900% and improving content accuracy by 95% through dynamic prompt classification and LLM routing. • Integrated Chainlit for real-time conversational AI, boosting user engagement by 70% and streamlining documentation requests, resulting in a 5% reduction in support tickets. • Reduced model retraining time by 95% by integrating LangChain for optimized LLM workflows, ensuring faster, context-aware documentation updates across multiple integration pipelines.
• Developed and deployed efficient machine learning models, achieving 4x smaller size weights, 500x energy efficiency, and 4x faster processing for real-world applications, emphasizing performance and sustainability. • Led a multimodal anomaly detection project that integrated sensor data (vibration, pressure, temperature, flow rate, voltage, and current consumption), enhancing detection accuracy across diverse domains, and showcasing adaptability and innovation. • Managed end-to-end collaboration with hardware firms, especially Zalmotek, to successfully build and demonstrate the remarkable capabilities of BrainChip's Neuromorphic platform, underscoring my pivotal role in driving the development and delivery of cutting-edge anomaly detection solutions. • Spearheaded the development of distracted driving technology, achieving energy and processing gains, positioning for potential project collaboration with Mercedes, and showcasing the capabilities of the AKD1000 in the automotive safety domain.
AI4ALL is a nonprofit working to increase diversity and inclusion in artificial intelligence. It creates pipelines for underrepresented talent through education and mentorship programs around the U.S. and Canada that give high school students early exposure to AI for social good. The vision is for AI to be developed by a broad group of thinkers and doers advancing AI for humanity’s benefit. • Planned and executed the summer programs teaching high school students about AI in collaboration with other faculty and staff. • Created a premier AI curriculum for high school students covering concepts such as Clustering, Classification, Naive Bayes, Regression, Neural Networks, Data and Bias. • Led two student group projects - Fake News Detection and Heart Disease Prediction covering various topics such as Data Preprocessing, PCA, Multinomial Naive Bayes, Gaussian Naive Bayes, k-Nearest Neighbors, Decision Trees, Random Forests, and SVM.