Monty Dean
Independent AI Developer
I am an independent AI developer and researcher, combining a strong foundation in Mathematics and Computer Science with hands-on expertise in building and deploying practical machine learning models. I specialize in making AI accessible and useful—from optimizing diagnostic summarization tools to engineering intuitive, high-performance machine learning solutions.
Certified in Azure AI Fundamentals (AI-900), I am seeking a professional opportunity where I can apply my analytical skills, rapid learning capability, and proactive problem-solving mindset to build impactful, real-world AI applications.
Show all 26 additional Codecademy certificates
Completed foundational coursework in mathematics, programming, and computer systems before pivoting to full-time independent AI research and development.
Mathematics (A), Further Mathematics (B), Computer Science (B)
Mathematics [9], Computer Science [8], Combined Science [6-6], Geography [6], English [6], English Literature [5], Music [Distinction]
- Answered the phone for a variety of companies on their behalf
- Booked appointments for clients
- Called and monitored the status of engineers for emergency situations
- Handled angry or challenging customer scenarios with professionalism
- Answered the phone for a variety of client companies
- Passed on detailed, accurate messages
- Handled customer complaints effectively
- Worked efficiently under time-sensitive call conditions
YieldStorage
A commercial stochastic dynamic-pricing SaaS for independent self-storage operators: nightly Bellman rate optimization, submarket competitor rate scraping, SiteLink/storEDGE/Yardi sync, and a fully client-side WASM revenue audit.
UK Public Utility Suite
An autonomously orchestrated WASM & TypeScript client-side suite protecting UK citizens from statutory fines, 62% tax traps, energy back-billing, and unlawful eviction notices.
Radiology AI Summarizer
A 2.97 GB clinical language model fine-tuned to compress dense findings into precise impressions on CPU (+156% ROUGE accuracy vs base).
Visual Product Recommender
An unsupervised computer vision recommendation system for clothing apparel using ResNet-50 2048-dim feature extraction & cosine similarity matching.
Spiral Data Classifier
A PyTorch-trained neural network demonstrating the power of adaptive optimizers (Adam vs SGD) with an interactive 2D canvas boundary plotter.