Available for interesting problems
I'm a Staff AI/ML Engineer at Genesys, architecting next-generation AI agents and LLM-powered applications. With 7+ years in software and 5+ years focused on NLP, machine translation, and generative AI - backed by a Master's in Data Science.
About
I design and deliver end-to-end machine learning systems - from data preprocessing and model training through evaluation, optimization, and production deployment. My recent work centers on generative AI: fine-tuning LLMs with PEFT methods, building agentic AI systems, and shipping scalable, production-ready applications.
Earlier in my career I spent five years deep in natural language processing and neural machine translation, building evaluation frameworks and language tooling that operated across 190+ languages. That mix of research depth and engineering pragmatism is what I bring to hard AI problems today.
Outside of engineering you'll find me reading, playing basketball, and keeping up with the relentless pace of AI research.
Fine-tuning, prompting, and productionizing large language models for real workloads.
Orchestrating multi-step agentic systems and retrieval pipelines with LangGraph and MCP.
Scalable, efficient serving - quantization, ONNX, and clean FastAPI system design.
Career
Genesys · Dublin, Ireland
R&D / Data Science Group. Architecting next-generation AI agents and LLM-powered applications.
Oracle · Dublin, Ireland
Neural machine translation, LLM fine-tuning, agentic systems, and evaluation frameworks.
Oracle · Bangalore, India
Cross-platform application development across globally distributed teams.
Oracle · Bangalore, India
Core product feature development with a focus on performance optimization.
Trinity College Dublin · Dublin, Ireland
Specialization in machine learning and artificial intelligence.
Amrita School of Engineering · Coimbatore, India
Foundations in data structures, algorithms, and systems.
Toolbox
Oracle · 2023 · PDF
Oracle · PDF
Work
2025
A conversational agent that answers natural-language movie questions over the MovieLens dataset, built as a multi-agent LangGraph workflow with SQL tool-calling, an MCP weather server, and a Streamlit chat UI.
2024
A Proxmox-based home lab running around two dozen self-hosted services as version-controlled Docker Compose stacks, with a reverse proxy, local DNS, VPN access, automatic updates, and scripted backups.
2020
A CNN-based captcha breaker that generates image and audio (text-to-speech) captchas, trains a multi-head convolutional neural network on them, and classifies unseen captchas.
2020
A Python and pygame implementation of Connect 4 with five selectable AI bots (random, one-step look-ahead, Minimax, Expectimax, and Monte Carlo Tree Search) that can play a human or each other, plus a win-ratio evaluation across the agents.
2020
A Flappy Bird AI that teaches itself to play using NEAT (NeuroEvolution of Augmenting Topologies), evolving small neural-network controllers over successive generations.
Publication
7th ICML Workshop on Automated Machine Learning (AutoML) · June 2020
Contact
Have an interesting AI/ML problem, a role, or just want to compare notes on agents and LLMs? The fastest way to reach me is email - I read every message.
Based in Dublin, Ireland