Software engineer and physics researcher building GPU-accelerated AI systems, scientific tooling, and polished products.
Machine learning is my primary field of interest.
PINN Simulation
Carbon nanocluster system
@tf.function(jit_compile=True)
def loss_fn(model, x, t):
u = model([x, t], training=True)
residual = pde_residual(u, x, t)
bc = boundary_loss(u, x, t)
return residual + 0.2 * bcTraining Progress
LiveEdge AI Preview
CoreML inference - Apple Watch Series 9
Inference
0.98
iPong
Real-time policy
128 FPS
80%
PINN accuracy lift
on carbon nanocluster predictions
60%
faster simulations
with a GPU-accelerated pipeline
2.88M+
AR views
across Snapchat Lenses
Department of Physics, King's College London
Applied physics-informed neural networks to model carbon nanoclusters, boosting prediction accuracy by 80% and cutting simulation time by 60% with an end-to-end TensorFlow + CUDA pipeline.
Jun 2025 - Present
Kannan Industrials
Architected and deployed iOS applications including 1minute DOEShelp, iPong, and DabCounter. Integrated CoreML for real-time inference, improving speed by 70% on Apple Watch hardware.
Feb 2021 - Present
Kennedy Institute of Rheumatology
Developed CNNs for biomedical imaging with a 40% improvement over baselines, and implemented reinforcement-learning systems using LSTMs with mixed-precision GPU training.
Nov 2019
NHS Digital
Built an LSTM + Word2Vec NLP system to auto-assign ICD-9 codes from clinical notes, raising accuracy from 42% to 71% while adding anomaly detection tooling.
Nov 2019
featured project
# LAMMPS input excerpt
units metal
atom_style atomic
boundary p p p
pair_style eam/alloy
neighbor 2.0 bin
fix 1 all nvt temp 300 300 0.1
run 500000
Parallelised molecular simulations for material discovery using AIRSS search strategies with LMP KOKKOS, OpenMP, and CUDA acceleration.
Bridging physics, machine learning, and real-world product constraints with systems that can be measured, tuned, and shipped.
Physics-informed models for materials and molecular systems.
Real-time inference on constrained Apple devices.
Scalable systems from prototype to production.
BSc Physics - Passed Year 1
2024 - 2027
King’s Undergraduate Research Fellowship 2025. Recipient of the Alessandro de Vita Computational Physics Prize 2024-25. Relevant modules include Mathematics & Computation, Classical Physics, Astrophysics, and Modern Physics.
MEng Computer Science
2021 - 2023
Focused on algorithms, machine learning, theory of computation, distributed systems, and software engineering.
A Levels
2012 - 2019
A*A*A* in Mathematics, Further Mathematics, and Physics. 7 A*s and 4 As at GCSE.
Languages
AI / Machine Learning
Cloud & DevOps
Systems & Tools
Public LinkedIn profile details add a fuller picture of the mobile, UX, cloud, and machine-learning work behind the portfolio.
Meta
Issued Nov 2025
Meta
Issued Sep 2025
Meta
Issued Sep 2025
Meta
Issued Sep 2025
Meta
Issued Sep 2025
DeepLearning.AI
Issued Mar 2021
Stanford University
Issued Mar 2021
Nov 2025
King's College London
Recognizes participation in the King’s Undergraduate Research Fellowship programme after three months of summer research.
Nov 2024
King's College London NMES
Awarded for completing first-year computational challenges requiring original programs for code-breaking and biophysics tasks.