I'm Devshree — a computer science graduate with a passion for machine learning, data analytics, and building full-stack applications that turn complex data into clear, actionable insight.
Advanced Comp Sci (Distinction)
Electronics & Telecom · India
Portfolio Projects
Who I Am
Computer science graduate passionate about machine learning, data analytics, and building full-stack applications. Based in the UK, open to opportunities.
I'm Devshree Deshmukh, a computer science graduate with a passion for machine learning, data analytics, and building full-stack applications that turn complex data into clear, actionable insight.
I hold an MSc in Advanced Computer Science (Distinction) from the University of Leicester and a BTech in Electronics & Telecommunication from India. My MSc dissertation on AI-based medical imaging analysis earned a Prize Certificate.
I'm based in the UK and open to opportunities in data science, ML engineering, and software development. I love turning raw numbers into compelling data stories and building end-to-end systems that solve real problems.
Building predictive models, classification pipelines, and segmentation systems with a focus on interpretable outputs.
Wrangling large datasets, building SQL pipelines, creating dashboards, and communicating insights to non-technical audiences.
Creating end-to-end web applications — from Flask and REST APIs on the backend to interactive React frontends.
Exploratory data analysis, statistical visualisation, and turning raw numbers into clear, compelling data stories.
Where I've Worked
Professional experience applying machine learning and data engineering to solve real-world problems.
Developed predictive ML models to optimise business KPIs, improving model accuracy by 15% through feature engineering. Built automated SQL + pandas pipelines processing 5M+ records/week. Collaborated with cross-functional teams to translate requirements into model features and deliver stakeholder-ready outputs.
Developed a full-stack AI web application for MRI tumour analysis featuring EfficientNet-B0 classification, Attention U-Net segmentation achieving 95%+ Dice Score, 3D visualisation, and downloadable reports. Awarded Distinction and Prize Certificate.
Tech Stack
The languages, frameworks, and tools I use to build intelligent systems and data-driven applications.
Education
A strong foundation in computer science, electronics, and research-driven problem solving.
MSc Advanced Computer Science from the University of Leicester, completed with Distinction. My dissertation, MedSegment — an AI-based tool for brain tumour detection and medical imaging analysis — earned a Prize Certificate for its innovative approach combining deep learning classification, attention-based segmentation, and a full-stack web application.
Selected Work
Real projects I built — from AI-powered web apps and data dashboards to exploratory analyses and computer vision pipelines.
Full-stack AI web app for MRI tumour analysis — EfficientNet-B0 classification, Attention U-Net segmentation achieving 95%+ Dice Score, 3D visualisation, and downloadable reports. MSc Dissertation — Distinction & Prize Certificate.
Source Code →Interactive Tableau dashboard analysing churn patterns across 10,000 European bank customers. Identified 20.4% churn rate, Germany as the problem market (32.4%), and a product overselling risk driving attrition.
Source Code →EDA of 1,500 AI job listings across 14 countries with 16 statistical visualisations. Key insight: Python in 62.8% of listings, MLOps fastest-growing at 55.4% YoY, LLM roles at $191k median salary premium.
Source Code →CNN-based fashion classification pipeline achieving 85% accuracy across 10,000+ samples, with weather API integration for context-aware recommendations and an interactive dashboard.
Source Code →Coming Soon
Currently working on an end-to-end data analytics case study with real-world UK rail data.
End-to-end data analytics case study using real ORR rail performance data — exploring delay trends, operator rankings, cancellation patterns, and passenger impact with a custom disruption scoring metric. Includes interactive Tableau/Power BI dashboard and future ML extensions for delay prediction.
Open to Opportunities
Get in Touch
Have a project in mind, a question, or just want to say hello? I'd love to hear from you.
Whether you're a company looking for an ML engineer, a team in need of data-driven insights, or just want to chat about tech and AI, I'd love to hear from you.
devshree2201@gmail.com