I’m an AI Engineer working across computer vision, multimodal AI, generative AI, and the application stack needed to turn models into usable software.
About
My work spans computer vision, multimodal AI, and generative AI, with an engineering stack that covers model development, APIs, data systems, application interfaces, and deployment.
Recent work includes an oral-lesion classification prototype with interpretable predictions and portable offline inference, and a multimodal deepfake detector evaluated across multiple public datasets.
Projects
Developed a multimodal deepfake video detector that combines visual and physiological signals across multiple temporal windows using SigLIP2, rPPG, CNN-Transformer encoding, and quality-aware cross-attention fusion. Built leakage-aware training and zero-shot evaluation pipelines across CelebDF v2, FaceForensics++ C23, and WildDeepfake.
95.73%In-Domain Accuracy
79.10%Cross-Dataset Macro
3Datasets Evaluated
Designed and delivered an end-to-end computer vision prototype for binary classification of oral lesions as benign or malignant using a fine-tuned DINOv2 and EfficientNetV2-B3 ensemble. Added Grad-CAM explanations for qualitative review and packaged the system as a PyQt6 desktop application that runs on clinic hardware without a network connection.
0.96Test AUC-ROC
ONNXPortable Inference
OfflineRuns On-Device
Skills
Python, TypeScript, SQL.
LLM APIs, retrieval-augmented generation, embeddings, retrieval and reranking.
Structured outputs, tool calling, AI agents, LLM evaluation and observability, multimodal AI.
PyTorch, YOLO, OpenCV.
React, Next.js, FastAPI, PostgreSQL, pgvector, Redis, Celery.
Docker, Vercel, AWS, GitHub Actions.
Contact
Find my work on GitHub or connect with me on LinkedIn.