AI & ML Engineer · Researcher · Limerick, IE

Enda
O'Shea

I build production AI systems — from RAG pipelines and multi-agent LLM platforms to computer vision and classical search. My work spans academia and industry, with a Ph.D. focused on applied AI.

Dr. Enda O' Shea
Ph.D.
University of Limerick
M.Sc.
University of St. Andrews
M.Sc.
Maynooth University
B.Sc.
University of Limerick
01 — 7 works

AI Engineering

Production LLM systems, RAG pipelines, multi-agent platforms. Each shipped behind real traffic with stateless services, instrumented endpoints, and evaluation harnesses.

LLM API Integration Microservice
Featured

LLM API Integration Microservice

An MCP server that gives any AI coding assistant instant access to live model discovery, pricing comparison, and ready-to-use connection snippets across seven providers — Anthropic, Google, OpenAI, DeepSeek, Z.ai, MiniMax, and Inception Labs. Built as a stateless HTTP microservice behind nginx, it plugs into Claude Code, Gemini CLI, and OpenAI Codex with a single URL — no API keys, no SDK installs, no configuration drift. One endpoint turns "which model should I use?" into an answered question with working code in five languages.

200+
Models
7
Providers
5
Languages
MCP (Model Context Protocol)REST API DesignCI/CD (GitHub Actions)Plugin ArchitectureGraceful DegradationMulti-client Interoperability
Live

RAG Decision MCP

A Docker-first MCP server that gives AI agents structured, explainable guidance for building RAG systems — which architecture fits a workload, how to chunk each modality, and what retrieval stack, vector store, and evaluation setup to use at a given scale. A curated, versioned knowledge base drives deterministic, read-only recommendations with typed contracts and Mermaid architecture visualizations — no scraping, no runtime LLM calls. Hosted publicly: one URL, no API key, works identically across Claude Code, Gemini CLI, and OpenAI Codex.

23
RAG patterns
26
Retrieval techniques
18
Vector DBs
22
Eval metrics
MCP (Model Context Protocol)TypeScriptNode.jsZodKnowledge EngineeringDockerNginx
The Architect
Live

The Architect

Systems Architect is a browser-based studio for system design interviews. It pairs a drag-and-drop canvas — built around a 7-layer architecture model with concrete tool choices at every layer — with live, realtime interview sessions: interviewers invite candidates by link, watch them design in real time, and pass canvas control back and forth. The AI layer (Gemini, OpenAI, or Claude — your key, your provider) generates calibrated challenges, scores designs against a six-dimension rubric, publishes hints, and walks candidates through a model solution step-by-step. Solo practice and live interviews share the same canvas, so what you rehearse alone is exactly what you run with a candidate.

ReactTypeScriptViteCanvas APIGemini APINginxComponent-Based ArchitectureGA4
ML Interactive Labs
Live

ML Interactive Labs

Interactive machine-learning education platform spanning nineteen subject areas — from the original Policy Playground (Q-Learning, SARSA, bandits, multi-agent RL) through classic ML, search, deep learning, and Bayesian models. Every lab is a live, analytic, client-side simulation with real-time mathematical analysis, wrapped in a full-screen Cinematic Stage UI with a docked multi-provider AI tutor (Gemini, OpenAI, Anthropic, or DeepSeek — bring your own key). No backend, no servers.

Reinforcement LearningReactTypeScriptViteMulti-provider LLM APIsNginxWeb Crypto APIRate LimitingDocker
Document Interaction System
Archive

Document Interaction System

RAG-powered Q&A system using Docling for multi-format document extraction, LangGraph agents with Google Gemini 2.5 Flash, and ChromaDB vector search. Dockerized with GPU acceleration, featuring markdown-aware chunking and streaming responses for intelligent document querying.

PythonLangGraphChromaDBOCRPyTorchCUDADockerStreamlit
Neural Refresh
Live

Neural Refresh

A React-based CS knowledge reinforcement app that helps developers combat skill atrophy through curated trivia questions across 10 domain categories, with optional AI-powered personalized learning plans via Gemini API.

LLMsGenerative AIGemini APIClaude CLIReactDockerVPS Deployment
Prompt Generator and Evaluator Agent
Live

Prompt Generator and Evaluator Agent

Web-based multi-provider prompt-engineering workbench. Generates platform-optimized prompts for Claude, ChatGPT, or Gemini and refines them with a dual-agent evaluator-revisor loop that prevents self-evaluation bias. Runs on Google, OpenAI, Anthropic, or DeepSeek engines with bring-your-own API keys held only in the browser session, selectable evaluator engines, and per-request thinking-level control.

Prompt EngineeringLLMsMulti-Agent SystemsBYO API KeysGemini APIOpenAI APIAnthropic APIDockerVPS Deployment
Now building

In active development — links go live as soon as they're ready.

In development

Founder OS

An evolution of the Systems Architect studio into a full product: user accounts, per-user project libraries, and designs stage-tracked from hobbyist to enterprise scale across a six-provider AI layer.

In development

Sports Management System

A full-stack team management platform for GAA clubs — squad rosters, drag-and-drop tactical pitch planning, match scheduling, player wellness tracking, and end-to-end encrypted team messaging, shipping as a web app plus native iOS/Android builds via Capacitor.

In development

Isteach

A full-stack relationship management app for couples — shared calendar, self-hosted AI voice journaling, photo memories, date planning, and budget tracking, wrapped for iOS with Capacitor and in TestFlight.

Live demo ↗
In development

Tuig

A daily-curiosity iOS app — one shared subject a day with three honest angles on it, served from a deterministic, backend-free content pipeline; in TestFlight, App Store submission in prep.

Live demo ↗
In development

k-q Roundtrip

An ongoing mechanistic interpretability study of GGUF k-quantization using a "dequant round-trip" method: quantized weights are dequantized and loaded into a hookable PyTorch model, making real-world deployment quants directly measurable with interpretability tooling for the first time.

Engage

Get in
touch.

Ask about my work