Nicholas Fidalgo
Harvard EECS showcasing my skills in software, hardware, AI and machine learning
Education
M.S. Computer Science B.A. Electrical & Computer Engineering
Harvard University
August 2022 – May 2026
Affiliations: ColorStack, Harvard Data Analytics Group (HDAG), Harvard Tech for Social Good (T4SG), Hispanic Scholarship Fund (HSF Scholar, Youth Leadership Institute)
Experience
Discover my expertise in software, hardware and machine learning developing impactful healthcare innovations.
Massachusetts General Hospital
Pre-processed patient data, reducing dimensionality by 86% while retaining 90% of information using PCA, standardization, and imputation techniques in Python. Improved training accuracy by 35% through feature engineering and developed a polynomial regression model with 94% accuracy to assess the impact of GLP-1 agonists on multiple sclerosis progression, identifying no significant relationship. Co-authored a peer-reviewed publication in Neurological Sciences, contributing to clinical decision-making and research in MS treatments.
August 2023 – Present
Data Scientist Intern
Harvard Ability Lab
Engineered a hardware control system for an automatic navigation aid for the visually impaired, resulting in a prototype capable of real-time obstacle avoidance and reducing navigation errors by 87%. Developed a PID control algorithm in C++ and integrated motor control with obstacle data using Bluetooth communication in Unity, enhancing the system’s responsiveness. Conducted rigorous testing through a maze navigation experiment, optimizing the system’s performance and reducing navigation time by 55%. Collaborated with a multidisciplinary team on three prototypes, demonstrating the system's feasibility.
September 2023 – Present
Software & Hardware Engineering Intern, Bioengineering Research
Projects
Below are some of my most recent projects and links to each of them.
Computer Vision: Structure from Motion
Developed a Python-based structure from motion algorithm, using RANSAC to improve 3D reconstruction accuracy by 44%. Computed essential matrix and camera positions, triangulated 3D points, and explored Tomasi-Kanade matrix factorization for enhanced results.
FPGA Design: MIPS Processor
Programmed a multicycle MIPS processor on a Xilinx FPGA using Verilog in Vivado, executing R, I, and J-type instructions. Designed a custom clock that optimized processing speed by 61%, and developed the full MIPS processor datapath and ALU.
FPGA Design: Cache Architecture
Simulated different cache architectures in Verilog using Vivado, including direct-mapped, fully associative, and set-associative caches. Built an accompanying Python assembler to facilitate testing and validation of each architecture, ensuring accurate performance analysis across various configurations.
FPGA Design: Traffic Light Controller
Developed and implemented a traffic light controller on a Xilinx FPGA using finite state machines (FSM) and timing logic, successfully demonstrating the functionality with a live demo on the FPGA board.
Machine Learning: 2024 Copa America
Predicted 2024 Copa America outcomes by web scraping match statistics, cleaning data with imputation and PCA, and training multiple machine learning models. Achieved 66% test accuracy in predicting win, loss, or draw using logistic regression, decision tree, random forest, bagging, AdaBoost, and gradient descent models.
Enhancing Digital Comm. Systems
Enhanced digital communication systems by implementing and simulating various encoding, modulation, and error correction techniques, including BPSK, QAM, repetition coding, and Hamming codes. Developed a Python file to simulate and analyze communication protocols and their application in overcoming real-world challenges, such as those encountered by the Galileo probe.
Skills
Languages
Python, Java, C++, SQL, JavaScript, Go, R, HTML, CSS, Verilog, VHDL, English, Spanish (intermediate)
Frameworks
React, Node.js, TensorFlow, PyTorch, scikit-learn, NumPy, pandas, Flask, Django, Express.js
Technologies
Git, Docker, AWS (S3, EC2), Google Cloud Platform, Azure, PostgreSQL, MongoDB, Jupyter, Firebase