Applied AI Prototyper · Creative Technologist · Web Platform Builder

I build tools that turn complex technical capabilities into useful, understandable products.

My work sits at the intersection of applied AI, semantic exploration, media processing, creative tooling, and web platforms. I’m especially interested in research prototypes and interfaces that help people inspect, navigate, and make practical use of model outputs.

Professionally, I work at TechSmith across web platform strategy, WordPress development, analytics, SEO, accessibility, experimentation, and applied AI.

An experimental interface for sampling and exploring spaces of generative-model responses.

It combines embeddings, lexical features, dimensionality reduction, clustering, nearest-neighbor search, run comparison, and interactive visualization to make patterns within model outputs easier to inspect.

Built with: Python, FastAPI, React, TypeScript, UMAP, HDBSCAN, KMeans, sentence embeddings

A local-first video preparation and optimization application for producing browser-ready media variants.

The project explores codec selection, quality estimation, subtitle workflows, packaging, job orchestration, reusable presets, and practical FFmpeg automation through a web interface.

Built with: TypeScript, React, Node.js, FFmpeg, FFprobe

A retrieval and evidence-exploration system for working with collections of research papers.

It includes document hydration, chunking, embeddings, FAISS retrieval, source-diversity controls, quote extraction, evidence-aware generation, and configurable multi-stage retrieval.

Built with: Python, FastAPI, React, embeddings, FAISS, retrieval-augmented generation

An image color-analysis and palette tool focused on usable design output rather than simple dominant-color extraction.

It includes perceptual color conversion, clustering, contrast and accessibility analysis, harmony generation, and interactive palette exploration.

Built with: Python, FastAPI, React, TypeScript, scikit-learn, CIELAB, WCAG contrast analysis

A browser-based creative video tool that manipulates H.264 NAL units to produce configurable datamoshing and compression-artifact effects.

The implementation works directly with encoded video structure, including IDR frames, SPS/PPS units, frame duplication, corruption, dropping, and reordering.

Built with: Python, Flask, FFmpeg, H.264 bitstream processing

• Research prototyping and internal research tools
• Applied AI and retrieval systems
• Human-AI interaction
• Semantic-space and model-output visualization
• Multimodal and creative AI workflows
• Media-processing and browser-delivery systems
• Accessible, usable interfaces for technically complex systems

I tend to begin with an ambiguous problem, build an end-to-end prototype, and use the result to discover what the product or research question should become.

My strongest work usually involves:

• Translating broad ideas into functioning systems
• Structuring model outputs for downstream workflows
• Connecting AI capabilities with real product interfaces
• Combining unfamiliar technologies quickly
• Designing around the complete user workflow rather than an isolated feature

I’m currently focused on turning ambitious prototypes into more modular, tested, measurable, and reproducible systems.

Languages: TypeScript, JavaScript, Python, PHP, SQL, HTML, CSS

Platforms and frameworks: React, Node.js, FastAPI, Flask, Streamlit, WordPress, Gutenberg, MySQL, Linux

AI and data: OpenAI APIs, structured generation, embeddings, RAG, FAISS, sentence-transformers, UMAP, HDBSCAN, KMeans

Media and web: FFmpeg, FFprobe, video codecs, accessibility, technical SEO, GA4, GTM, Cloudflare