- Support Cornell's foundations of artificial intelligence course (CS 4700)- Instruct students one-on-one during weekly office hours to help them develop skills ranging across knowledge representation, heuristic search, problem-solving, natural-language processing, game-playing, logic and deduction, planning, and machine learning- Aided in developing instructional material and refining exam questions

- Implemented maps visualization tool working in tangent with data transformer pipeline in Java to generate polygons of incomplete cloud lanes making up ~97.86% of all cloud lanes by deriving geometry based on segment polylines and default lane widths and orderings, aiding internal and external debugging on automotive cloud API platform. - Developed and automated PLX dashboard as source of truth for debugging inconsistencies data from automotive maps API, boosting on-call efficiency of future engineers by 2x and eliminating need to manually search several different sources of data for information.- Full development, using Flume, protocol buffers, configuration files, SQL, Spanner database, Midas Package Manager, and other internal structures- Followed up with additional investigation of storage and runtime for all projects to check if additional resources need to be reallocated for future use.

- Created an ETL pipeline in Java for reading, writing and processing info between external sources and HR API, using Apache Beam SDK and internal Google infrastructures to maintain data security and scalability.- Designed and implemented code auto-generation tool to speed up code-writing process by 2x for next 90+ pipelines, allowing future engineers to focus more on implementing business logic transforms instead of constructing basic framework of the pipeline.- Backend development, using XMLs, protocol buffers, Spanner databases, SQL, configuration files, decryption/encryption systems, and other internal structures.- Completed entire development process, including considering several structure designs, writing design docs accordingly, completing full implementation, reprioritizing tasks around blockers, going through design review, end-to-end testing with JUnit and manual testing, and launching internally.

- Designed packages for Raman spectroscopy data analysis and image analysis in disease identification for stimulated Raman scattering microscopy, such as high dimensional reduction analysis, peak deconvolution and quantitative sub cellular morphological analysis. - Published “Direct Imaging of Lipid Metabolic Changes in Drosophila Ovary During Aging Using DO-SRS Microscopy” and co-authoring other papers on methionine regulations in lifespan and quantification of dietary affects on survivability in 400+ drosophila.- Compiled a publication paper about lipid metabolism in Type 2 Diabetes, comparing 80+ spectrums and 100+ SRS images produced from athletic, lean, obese and diabetic muscle tissues.

- Developed software to compile images into 3D structures to computationally obtain more accurate measurements for image analysis on cell cultures.- Conducted wet-lab experiments on cell cultures with overnight monitor of nuclei responses to environmental stressor using multi-photon microscopes and corresponding ZEN software.- Identified specific proteins in nuclear lamina and tunnels that respond to certain stressors and alter chromatin for DNA transcription.