Royal Holloway, University of London
Graduated with First Class Honours and received the University of London Driver Prize for the highest grades in the graduating year.
About
Astrophysicist turned AI leader, still driven by first principles.
Darsh Kodwani works across machine learning, scientific thinking, and business execution. The through-line is a conviction that rigorous, data-driven technology can help solve the world’s biggest problems.
Current role
Senior Data Science Manager at Microsoft, focused on AI solutions for asset-heavy industries.
Background
Born in Ahmedabad, raised largely in Adipur, moved to the UK at 12, with formative academic years in Toronto and Oxford.
3Degrees from Royal Holloway, Toronto, and Oxford
9Peer-reviewed publications across cosmology, gravity, and ML
10+Years spanning research, startups, consulting, and enterprise AI
Narrative
What shapes the work
Darsh was born in Ahmedabad, Gujarat, and spent most of his childhood in Adipur in western India, with a brief early period in Manchester. He moved to the UK when he was 12 and has lived there since, except for two years in Toronto during his master’s degree.
His day-to-day work blends machine learning, software engineering, and strategic problem-solving. Alongside that, he continues to write and think about research questions in physics, statistics, and machine learning.
Statistics, mathematics, and programming are the common languages behind most of his work. Many of those tools were sharpened outside the classroom through self-learning, industry projects, and teaching.
Learning and teaching remain central themes. That same curiosity drives his broader view that rigorous technical work should ultimately create useful real-world outcomes.
Education
Academic foundations
University of Toronto
Graduated with a GPA of 3.92/4.0 under the supervision of Ue-Li Pen, strengthening the bridge between theory and computation.
University of Oxford
Completed doctoral research under Pedro G. Ferreira. Thesis: Origin and evolution of the universe.
Royal Holloway
Toronto
Oxford
Experience
Work across science, startups, consulting, and enterprise AI
Microsoft
Leads AI work for industry solutions in the UK, especially for asset industries. The work spans predictive maintenance, remaining useful life models, and cloud-scale software deployment across cross-functional global teams.
Accenture
Led teams of data scientists and ML engineers building analytics and AI tools for asset-industry clients, while also contributing to internal ML operations capabilities inside the Institute of AI.
Oxford Nanoimaging
Worked on analysis and automation for biological microscopy data at a biotech company built around super-resolution imaging technology from Oxford Physics.
EcoSync
Built supervised learning and reinforcement learning systems to optimize energy usage in buildings, combining practical product thinking with climate-tech impact.
Outsmart Insight
Delivered scientific and technical R&D consulting projects ranging from neuromorphic computing forecasts to deep-tech due diligence and expert-led market analysis.
Oxford Strategy Consulting
Led Oxford student and researcher teams on client work in logistics and clean energy, translating ambiguous client goals into structured project execution.
Earlier roles
Internships and early research work
Pythia Sports
Worked on supervised machine learning models for football analytics using large-scale match datasets and feature engineering.
National Physical Laboratory
Developed novel alloy-based temperature calibration approaches for thermocouples inside experimental apparatus.
Open University
Automated the opening and closing of the PIRATE telescope in Mallorca based on weather analysis, earning a Gold Crest award from the British Science Association.
Capabilities
Technical strengths
Python
Machine Learning
Statistics
Git
SQL
HTML / CSS
CI / CD
Rust
JavaScript
Docker
Contact
Continue the conversation
For research collaborations, AI consulting, speaking, or hiring conversations, the fastest routes are LinkedIn, GitHub, Google Scholar, or direct email.