Selected work

Generalist software engineer building practical systems across AI, backend and deployment.

Working across applied AI, computer vision, backend development and deployment workflows.

View featured projects Get in touch

Selected work

Applied AI, workflow products and analytical simulation.

Three projects showing different ways of building useful software.

2026Research project

Autonomous Lunar Lander with DQN

A Deep Q-Network agent trained to land autonomously in LunarLander-v3.

Applied AI / Reinforcement Learning

Designed and evaluated a DQN agent for autonomous landing in LunarLander-v3. The project covers problem framing, algorithm selection, PyTorch implementation, replay buffer and target network training, baseline comparison, multi-seed evaluation and ablation analysis.

A compact benchmark for learning decision-making from reward signals, useful when control policy cannot be hand-coded reliably and the system has to improve through trial, feedback and evaluation.

Applied AIReinforcement Learning

Stack

Python · PyTorch · Gymnasium · YAML

2026Live

Persona — AI-powered newsletter assistant

An AI-assisted news workflow with user preferences, chatbot interaction and scheduled newsletter delivery.

Built and deployed an AI-assisted news personalization workflow with a public web interface. Persona captures user preferences, aggregates stories from multiple sources, filters relevance and supports scheduled newsletter delivery by email.

Applied AIAutomationProduct

Product / automationn8n · Web interface · Email automation

Visit project

2026Personal project

Interstonar

A personal analytical and simulation project combining planetary trajectory analysis with local geometric intersection tests.

Built a simulation-oriented command-line project with two analysis modes: a planetary trajectory model under Newtonian gravity using Euler integration, and a local geometry mode based on sphere-assisted ray marching and collision reasoning. The project focuses on numerical methods, spatial thinking and algorithmic implementation rather than product polish.

SimulationAnalytical Systems

Simulation / analytical systemsC · CLI tooling · Numerical simulation · Geometry

Applied AI
Computer Vision
Backend Development
Analytical Systems
Deployment Workflows

Approach

Practical systems, kept straightforward.

A generalist profile shaped by applied AI, computer vision, backend development and deployment work.

I build practical systems that combine implementation, experimentation and product-oriented interfaces. The work spans applied AI, computer vision workflows, backend features and deployment habits that make software easier to run.

Selected background

Applied reinforcement learning work includes a DQN agent for LunarLander-v3, with baseline comparison, multi-seed evaluation and ablation analysis.

Computer vision experience includes OpenCV-based experimentation, OAK-D Lite camera workflows and YOLO, including a food-recognition hackathon and a separate network coverage hackathon covering 4G / 5G maps, eligibility APIs and antenna placement scoring.

Laravel backend internship · July 2025 — December 2025Backend internship focused on Laravel development over a 6-month period, contributing to application features and production-oriented engineering work.

Deployment work includes DigitalOcean VMs, Caddy, SSH hardening, UFW, Git and container-based release flows.

Stack

A compact view of the tools I can defend honestly today.

Grouped by the kinds of work they support, instead of turned into a noisy badge wall.

Applied AI & vision

Tools and workflows used for experimentation, inference and image-based recognition.

Python
OpenCV
YOLO
OAK-D Lite
PyTorch

Backend & product

Web and application work backed by practical implementation choices.

Laravel / PHP
JavaScript / TypeScript
Next.js
n8n
Git / GitHub

Deployment & ops

Operational habits that keep small systems usable and secure.

Linux
Docker
DigitalOcean
Caddy
SSH security
UFW
Git

Contact

If you are building something that needs applied AI, backend discipline or careful deployment work, let's talk.

Email is the fastest way to reach me. GitHub and LinkedIn are available as well if that is easier.

Email me GitHub LinkedIn