A comprehensive collection of AI/ML projects showcasing real-world applications and innovative solutions
Real-time assistant that converts doctor-patient conversations into structured SOAP notes. Whisper-based system supporting long-form audio and noisy environments with custom speaker merging logic.
Agentic receptionist capable of patient intake, symptom triage, appointment scheduling, and FAQ answering. Custom intent-routing and memory persistence for multi-turn contextual dialogues.
System that ingests video, audio, and images to generate structured summaries or step-by-step processes. Uses Whisper, video captioning, and image understanding pipelines.
Automated CV pipeline that stages, furnishes, and remodels interior rooms via diffusion models. Integrated ControlNet for precise room-aware generation with segmentation and collision avoidance.
Embeddable AI assistant that reads website content, documents, and tables autonomously. Website-level RAG using DOM extraction, semantic chunking, and lightweight indexing.
Chatbot capable of answering complex property-related queries using contract PDFs, Excel sheets, and listings. Combined RAG with real-time web search for updated market trends.
Plug-and-play object detection + tracking library targeting developers, students, and educators. Provided wrappers for YOLO, MobileNet, and custom ONNX models with RTSP/IP camera support.
Sports analytics tool extracting per-player statistics using object detection + DeepSORT tracking. Generated player heatmaps, possession metrics, speed estimation, and event tagging.
Combined OCR + table extraction + layout analysis for field-level understanding of invoices & receipts. Automatic validation for totals, tax fields, vendor recognition, and duplicate detection.