Scientific software and simulation for photonics and semiconductor engineering

Who we are

Flax & Teal is a specialist software and scientific computing consultancy working at the intersection of physics, simulation, data science, and cloud infrastructure, with a strong emphasis on open, transparent, and sustainable software practices.

We collaborate with photonics, semiconductor, sensing, and advanced engineering teams to modernise complex scientific software and workflows, using open source where appropriate, while protecting intellectual property and preserving scientific integrity.

We regularly work with industry, research, and cluster partners across the UK, Ireland, EU, North America, and Asia.

Where we typically help

We are often brought in when simulation or modelling capability becomes a constraint rather than a differentiator:

• Scientific and optical simulation that no longer scales beyond individual machines
• Legacy solver code (Python, C++, Fortran, mixed stacks) that is difficult to extend or maintain
• On-prem or desktop tools that need secure cloud or HPC enablement
• Increasing demand for parallelisation, automation, and reproducibility
• Cautious exploration of ML or AI alongside physics-based models
• Internal R&D tooling that needs to evolve into a reliable engineering or product platform

Core capabilities

Scientific computing & simulation

• Optical, semiconductor, and physics-based simulation pipelines
• FEM, PDE-driven models, numerical methods
• Parallelisation using Python, Dask, Kubernetes, and HPC environments

Cloud & infrastructure for science

• Cloud-native orchestration for simulation workloads
• Secure, reproducible CI/CD for scientific software
• Cost-aware scaling for bursty and high-volume compute

Data, ML & AI (applied responsibly)

• Machine learning used to augment, not replace, physical models
• Validation-led proof-of-concepts
• Integration into existing scientific and engineering workflows

Product & platform enablement

• Turning internal tools into usable, maintainable platforms
• APIs, dashboards, and visualisation for non-expert users
• Knowledge transfer and long-term sustainability built in

Selected experience

Semiconductor optics simulation

Refactoring licensed, legacy simulation code into modular, cloud-ready workflows; enabling high-volume, bursty simulation across teams; estimated upwards of €60m savings through modularisation and optimisation.

Sensor and telemetry systems

Near-real-time ingestion and analysis of large-scale sensor data; secure, multi-tenant cloud infrastructure; ML-driven insights integrated into operational systems.

Scientific R&D collaboration

FEM-based electromagnetic modelling; HPC-enabled 3D simulation; reproducible, open scientific pipelines developed alongside domain experts.

Additional photonics and semiconductor case studies available.

How we engage

We typically work in focused, low-risk stages designed to build confidence early:

• Discovery and feasibility assessments
• Proof-of-concept simulation or refactoring work
• Architecture and technical roadmap definition
• Embedded collaboration alongside internal teams
• Ongoing support for critical scientific tooling

Our approach is collaborative, transparent, and grounded in practical delivery.

How we work

• Clear IP boundaries; client-owned code remains client-owned
• Open source used strategically, not ideologically
• No vendor lock-in
• Knowledge transfer and enablement built into delivery