|
I am a fourth-year undergraduate at Boston College studying Computer Science and Mathematics. I work as a research assistant at Prof. Donglai Wei’s Computer Vision lab, focusing on the area of connectomics. I lead the BC Machine Intelligence Group (club Discord), am a member of the BC Experimental Math and Machine Learning Lab, and compete as part of the BC Competitive Programming Team. Previously, I held a research internship at EPFL’s CVLab where I had the privilege of being mentored by Dr. Jiancheng Yang and Prof. Pascal Fua. Additionally, I interned at Emmerich Research Center under Dr. Eden Steven. Email / CV / Google Scholar / Github |
|
|
TriSAM: Tri-Plane SAM for zero-shot cortical blood vessel segmentation in VEM images |
|
Frenet-Serret Frame-based Decomposition for Part Segmentation of 3D Curvilinear Structures |
|
Generating Anatomically Accurate Heart Structures via Neural Implicit Fields |
|
Multiplexed Volumetric CLEM enabled by antibody derivatives provides new insights into the cytology of the mouse cerebellar cortex |
|
RibSeg v2: A Large-scale Benchmark for Rib Labeling and Anatomical Centerline Extraction |
|
Light storage and thermal-assisted switching of SrAl2O4:Eu2+,Dy3+ |
|
GrandChallenge dockerization Ported evaluation scripts from connectomics challenges to the GrandChallenge platform |
|
ChunkPipeline Created a Dask-powered framework for orchestrating data processing pipelines on massive-scale connectomics datasets; ported 3D algorithms (connected components, euclidean distance transform, and TEASAR skeletonization) from the Seung Lab to be chunking-compatible, allowing for efficient utilization of the Boston College Linux Cluster |
|
Mycelium culture contamination detection (2020) Developed and deployed algorithms for fungus contamination detection, used for the automated applicaton of hydrogen peroxide in the development of synthetic leather |
|
Palm oil fresh-fruit-bunch grading (2020) Standardized grading of palm oil fruit through XGBoost-based classification of color-based features for industrial use |
|
Black soldier fly instance segmentation (2020) Built an Arduino-based automated data collection system for larvae, used Mask R-CNN and YOLACT++ for instance segmentation; adapted DCNv2 to be compatible with NVIDIA’s APEX for memory-efficient training. Part of a larger project to utilize BSF larvae for waste management |
|
magicpickle simplifies the process of transferring pickled representations, making code which both accesses data stored remotely and visualizes it locally easy |
|
SonicSniffer (2023) A weekend project: attempted to reproduce A Practical Deep Learning-Based Acoustic Side Channel Attack on Keyboards (spectrogram to keystrokes) without hand-crafted feature extraction |
|
SimpleMMO PVP Tool (2021) Developed a web-based tool to allow SimpleMMO players to aggregate results of API endpoint “mining”. Built using a Flask/PostgreSQL/Bootstrap stack |
|
Angle dependent Ohmic-resistance modelling of twisted-bilayer graphene via sparse linear solvers |
|
Reversing nearness (2020) Gradient descent based solution to Al Zimmermann’s discrete optimization contest |
|
Vision-enabled robotic arm (2019) Rigged and programmed a 6-DOF arm augmented with (simple) object detection capabilities. Presented at Indonesia Science Expo 2019 |
|
Keyboards I am a huge proponent of split-ortholinear keyboards. I currently run Colemak-DH with Achordion on an Arch. dux |
|
Source adapted from Jon Barron's website |