projects
projects here focus on learning depth, not just finished products.
data processing tool
repoa python cli tool that ingests raw csv and json data, scrubs inconsistencies, filters and selects fields, generates summaries, and exports clean output - turning messy, untrusted data into something a system can actually use.
this is the foundation of any serious data pipeline. with more time, it grows into an automated etl system - scheduled ingestion, validation rules, database writes, and pipeline monitoring. the thinking here is the same thinking behind tools that process millions of rows in production.
cpu design in logisim + vhdl
repoa full cpu design built from the ground up - logic gates in logisim, mapped into synthesizable vhdl. covers the fetch-decode-execute cycle, register files, alu design, and control unit logic.
most software engineers never go this deep. this project is the starting point for custom silicon thinking - the same process behind designing fpga accelerators, asic chips, and embedded processors. eventually, this kind of work feeds directly into building hardware for robotics and ai inference at the edge.
neural ofdm channel estimation
repoa research-oriented wireless ai project using nvidia sionna to explore channel estimation in ofdm communication systems. it starts with classical least squares estimation as a baseline, then builds toward neural methods that can learn better channel behavior under noisy and resource-limited conditions.
this project sits at the intersection of machine learning, signal processing, and next-generation wireless networks. the goal is not just to simulate communication links, but to understand how ai can make wireless systems more reliable in low-resource environments - the kind of work that connects directly to 6g, edge ai, and connectivity challenges in emerging markets.
python for robotics foundations
repoa structured, evolving repo where i build python fundamentals with one goal in mind: applying them in ros, control systems, and robotic pipelines. not generic tutorials - robotics-contextualized experiments and notes i will actually reuse.
this is groundwork for autonomous systems. the patterns being built here - sensor data handling, state machines, timing logic - are the same ones that run in ros nodes, robot arms, and autonomous vehicles. what looks like learning now is architecture for future systems.
personal site + notes
opena minimal, self-built site that serves as a live record - projects, roles, and thinking, updated as things evolve. built with next.js and designed to stay out of the way.
more than a portfolio, it is a system for thinking in public. the notes section could grow into a technical writing archive, a knowledge base for robotics and ai concepts, or a newsletter - a long-term signal of how my thinking compounds over time.
more projects that shaped how i think about systems, software, and engineering.