CS & AI · University of Edinburgh · Healthcare AI & Drug Discovery

Victoria Paterson

Third-year CS & AI student at the University of Edinburgh. Portfolio spans the full early drug development pipeline: multi-label QSAR adverse effect prediction (ChEMBL, SIDER, Morgan fingerprints, SHAP), ODE-based tumour resistance simulation, and conditional VAE for inverse molecular generation. DAAD RISE scholar (publication pending). Passionate about building software that accelerates evidence generation and improves health outcomes.

Healthcare AI Cheminformatics QSAR ODE Modelling VAE Edinburgh, UK
View Projects GitHub

Projects

Drug discovery · AI for health
01 — Featured

QSAR Drug
Side Effect Predictor

End-to-end drug adverse effect prediction system for early-stage drug safety screening. Multi-label classifier trained on ChEMBL + SIDER across 15 adverse effect labels. Molecular featurisation using Morgan fingerprints (r2-r5) and 14 RDKit physicochemical descriptors. Weighted loss for class imbalance; atom-level SHAP interpretability for clinical explainability. Full-stack deployment with Flask REST API and JavaScript frontend.

15 AE labels · SHAP interpretability · Flask + JS

PyTorch RDKit scikit-learn SHAP Flask
GitHub ↗
02

TumorEvo: Resistance Dynamics

Four-compartment ODE model (sensitive, resistance A, resistance B, dual-resistant) modelling acquired resistance dynamics relevant to clinical treatment planning. RK4 numerical integration; four clinical dosing schedules (continuous, pulsed, metronomic, escalating); real-time growth/death flux analysis and resistance dynamics visualisation. Full-stack web application with interactive parameter controls.

NumPy SciPy Flask Chart.js ODE
GitHub ↗
03

Conditional VAE Molecular Generation

Conditional VAE trained on ChEMBL and conditioned on SIDER side effect profiles, enabling inverse molecular design: generating candidate molecules with desired adverse effect signatures. Extends the QSAR predictor into a generative direction, closing the loop between property prediction and molecular design. In progress.

PyTorch RDKit VAE ChEMBL SIDER
GitHub ↗
04

Deepfake Political Speech Detector

Multimodal AI system detecting manipulated politician videos. XceptionNet/EfficientNet for visual artefact detection (89-98% accuracy), Wav2Vec2 for audio deepfake detection (93-98% accuracy). Late fusion meta-learner combining modalities; target 96-98% combined accuracy.

PyTorch Transformers OpenCV Wav2Vec2 XceptionNet
GitHub ↗
05

ASL Gesture Recognition

Web-based ASL gesture recognition; ensemble of Random Forest (static) and LSTM (dynamic); 21-point landmark feature engineering with full WebSocket streaming.

MediaPipe OpenCV scikit-learn Flask WebSockets
GitHub ↗
06

Instagram Hate Comment Detector

Real-time hate speech detection model + Chrome extension for automatic comment blurring; 92% precision on test dataset.

TensorFlow JavaScript Chrome Extensions API
GitHub ↗

Experience

Research & industry
Summer 2025 RWTH Aachen University
DAAD RISE Scholar

AI Research Intern

  • Built AI-powered chatbot to parse and analyse complex policy documents across multiple countries
  • Developed algorithms to quantify inter-country policy coherence; created visualisation suite for policy direction analysis
  • Research findings currently undergoing publication in academic journal
NLP Transformers Python Research
2026 EDINA · UK National Mapping Service
University of Edinburgh

AI Software Engineering Intern

  • Full-stack LLM integration: Designed and built an LLM-powered conversational interface into a production web platform; implemented semantic search over structured metadata using embedding generation and vector retrieval
  • Web application development: Extended a production platform with new frontend features and backend API endpoints; built interactive data visualisation dashboards for stakeholder use
  • Data pipeline engineering: Developed Python automation scripts for batch processing and format conversion of large-scale structured datasets
  • GeoAI prototyping: Evaluated and deployed pre-trained LLMs (Ollama, OpenAI API); benchmarked model performance against domain-specific retrieval requirements
LLM Embeddings Flask React GeoAI
Present University of Edinburgh

Lab Demonstrator

  • Lead weekly lab sessions for Cognitive Science course; specialise in computer vision and NLP
  • Guide students through implementation of neural network architectures; consistently receive excellent feedback
Computer Vision NLP Teaching
2024 — Present Scottish Policy Forum · Labour Party

Technology Policy Advisor

  • Bring technical perspective to AI and technology policy discussions; collaborate with senior policymakers to draft position papers
Policy AI Governance
About Me.

I'm a third-year Computer Science and Artificial Intelligence student at the University of Edinburgh with a focused portfolio in healthcare AI and computational drug discovery. My projects span the full early drug development pipeline: multilabel QSAR adverse effect prediction, ODE-based tumour resistance simulation, and conditional VAE for inverse molecular generation.

I complement this with a self-directed biomedical informatics curriculum covering biological APIs, sequence analysis, multi-omics, and biomedical NLP. I'm a DAAD RISE research scholar (publication pending) and have production full-stack software deployment experience.

I speak English, Norwegian, and German. Looking for opportunities in AI/ML for healthcare, drug discovery, or biomedical AI research.

Get in touch

Skills

Tech stack
CheminformaticsRDKit, Morgan fingerprints, QSAR
Healthcare AIChEMBL, SIDER, SHAP, ODE
Deep LearningPyTorch, TensorFlow, VAEs
NLPTransformers, spaCy, NLTK
Computer VisionOpenCV, MediaPipe, XceptionNet
BackendFlask, Node.js, REST APIs, WebSockets
FrontendReact, JavaScript, HTML/CSS
DatabasesMongoDB, PostgreSQL
DevOpsGit, Docker, Linux, CI/CD, AWS
LanguagesPython, JavaScript, Java, TypeScript, C#, Haskell
Get in touch

Let's work together.

victoriaflora2005@gmail.com