Prompt to Perception
Published:
Developed an end-to-end system for text-to-image generation using prompt refinement and Stable Diffusion 2.1. Improved prompt quality 3× (ROUGE metrics) via T5-Small, with an average prompt-image alignment of 0.72
Posts by Collection
portfolio
projects
Diseased Plant Detection
Published:
Comparative study of deep learning architectures (ResNet50, VGG16, and custom CNN) for plant disease classification across 38 different classes. Implemented and evaluated models on a comprehensive plant disease dataset, achieving state-of-the-art accuracy in disease detection and classification. Documentation available in the project repository.
CNN Visualization
Published:
Built an interactive system to inspect intermediate layers of CNNs. Given an input image, architecture, and PyTorch it renders per‑layer activation maps and optionally visualizes filter weights. Added forward hooks and epoch snapshots to step through training for real‑time introspection and debugging
Karma AI
Published:
An AI-powered accessibility solution that helps visually impaired people navigate their surroundings safely. Uses multimodal prompting (image + text) with Google Vertex API for scene descriptions and navigation assistance. Integrates OpenAI-based GPT voice assistant and fall detection using accelerometers for enhanced safety and independence.
TraitViz
Published:
Advanced depth estimation and 3D reconstruction using RGBD data from pinhole cameras. Implements state-of-the-art algorithms on the NYUD dataset.
narrAIt
Published:
narrAIt is an AI-powered accessibility companion for macOS that helps users understand complex interfaces through real-time guidance. It explains on-screen elements, answers voice questions in context, and points to exact interaction targets without taking over user control.
VoiceForge VR
Published:
VoiceForge VR is an interactive virtual reality project centered on intuitive in-world controls and user interaction flow.
Jarvis-home
Published:
Jarvis-home is my fully local AI assistant running on NVIDIA Jetson Orin Nano. It combines speech-to-text, LLM reasoning, and text-to-speech with an always-improving on-device pipeline, and I actively push upgrades to it almost every day.
publications
Paper Title Number 1
Published in Journal 1, 2009
This paper is about the number 1. The number 2 is left for future work.
Recommended citation: Your Name, You. (2009). "Paper Title Number 1." Journal 1. 1(1).
Download Paper | Download Slides | Download Bibtex
Paper Title Number 2
Published in Journal 1, 2010
This paper is about the number 2. The number 3 is left for future work.
Recommended citation: Your Name, You. (2010). "Paper Title Number 2." Journal 1. 1(2).
Download Paper | Download Slides
Paper Title Number 3
Published in Journal 1, 2015
This paper is about the number 3. The number 4 is left for future work.
Recommended citation: Your Name, You. (2015). "Paper Title Number 3." Journal 1. 1(3).
Download Paper | Download Slides
Paper Title Number 4
Published in GitHub Journal of Bugs, 2024
This paper is about fixing template issue #693.
Recommended citation: Your Name, You. (2024). "Paper Title Number 3." GitHub Journal of Bugs. 1(3).
Download Paper
Trustworthy LLM-Mediated Communication: Evaluating Information Fidelity in LLM as a Communicator (LAAC) Framework in Multiple Application Domains
Published in IEEE DISTILL 2025 (co-located with IEEE TPS 2025), 2025
The proliferation of AI-generated content has created an absurd communication theater where senders use LLMs to inflate simple ideas into verbose content, recipients use LLMs to compress them back into summaries, and as a consequence neither party engage with authentic content. LAAC (LLM as a Communicator) proposes a paradigm shift - positioning LLMs as intelligent communication intermediaries that capture the sender's intent through structured dialogue and facilitate genuine knowledge exchange with recipients.
Benchmarking Tabular Foundation Models for Agricultural Yield Prediction
Published in AgriAI 2026 Workshop (co-located with AAAI 2026), 2025
We benchmark TabPFNv2 against AutoGluon and PyCaret across three agricultural datasets and show that foundation models excel with missing or limited data, while AutoML dominates large clean datasets.
CRAFT: Corrective and Robust Multi-Agent Framework for Text-to-Parametric CAD
Published in ICLAD 2026 (IEEE International Conference on LLM-Aided Design), 2026
CRAFT turns natural language into editable, parametric CAD by combining semantic understanding, parametric planning, compilation, rendering, self-correction, and component-level verification within a layered recovery process, using multi-view visual feedback to detect and repair geometric errors.
teaching
COM S 336 (Computer Graphics)
Computer Science, Iowa State University, 2023
TA for COM S 336, an introduction to computer graphics. Students learned the graphics pipeline, coordinate systems, transformation matrices, and how to create interactive 3D applications. The course emphasized practical skills in OpenGL/WebGL, shader programming, and modern rendering techniques.
COM S 309 (Software Development Practices)
Computer Science, Iowa State University, 2024
TA for COM S 309, focusing on major software engineering topics and preparing students to lead medium-sized software projects. Emphasized teamwork, accountability, project management, and professional ethics through a semester-long group project.