Building Avkalan Labs
Building a technical venture around simulation software, custom numerical solvers, scientific machine learning, synthetic-data generation and engineering automation.
Role: Co-founder
View projectProducts, computational systems and engineering ventures.
Selected work across hardware-software products, simulation platforms, scientific machine learning, engineering algorithms and high-fidelity analysis.
.png)
.png)
.png)
.png)
.png)
.png)
.png)
Different engineering problems. One computational approach.
Numerical simulation, scientific ML, solver development, and engineering software for modelling physical systems and building usable technical products.
Project archive
Selected work
Explore the projects through product, venture, simulation, scientific machine learning and engineering-algorithm lenses.
EigenSole
An AI-powered smart wearable combining pressure and motion sensing, embedded electronics, gait analytics, dashboard software and customised insole development.
Role: Co-founder and product lead
View project
Synthetic-data pipeline for scientific machine learning
An automated pipeline for running physics-based simulation campaigns, converting results into structured datasets and supporting SciML training and validation.
Role: Project lead
View project
Underground-imaging SciML workflow
A physics-led workflow combining electromagnetic simulation, synthetic data, scientific machine learning and subsurface interpretation.
Role: Solver and scientific ML lead
View project
Thermal analysis algorithms for semiconductors and heat exchangers
Domain-specific algorithms for evaluating thermal behaviour, operating conditions and design choices in semiconductor and heat-exchanger systems.
Role: Engineering algorithm developer
View project
Ballistic-impact simulation and material response
A high-fidelity simulation study of deformation, material response and energy absorption under ballistic loading.
Role: Simulation and analysis lead
View projectRecurring themes
The domain changes. The operating questions remain similar.
Problem and user
What decision, workflow, or physical problem needs to improve and who depends on the result?
System representation
How should the physics, geometry, data, operating conditions and constraints be represented?
Technical trade-offs
What level of fidelity, speed, complexity and cost is appropriate for the intended use?
Evidence and trust
What validation, comparison, testing, or feedback is required before the result can be relied on?
Product and adoption
How does the technical capability become a repeatable, understandable and usable product or workflow?
Let’s talk about what you’re building.
I’m always open to conversations about engineering, computation, deeptech products and mentoring people building or learning in these areas.