Inverse Design
for Everyone

The world's easiest to use tool for inverse design. Leverage automatic
differentiation backed by Tidy3D's GPU-based FDTD solver to solve
your design problems effortlessly.

Start for free See a demo

TRUSTED BY More than 250 Companies & ACADEMIC INSTITUTIONS

Inverse Design
Reimagined

advantages

Simple Interface with GUI or PYTHON

Convert Simulation to Optimization in one line of code

Fast Speed

SIMULATION
EXAMPLES

Mode converter

Waveguide taper

Compact grating coupler

Bilayer grating coupler

Integrated bandpass filter

Metalens

Nanoantenna metasurface

Waveguide bend

Wavelength division multiplexer

Metalens-assisted waveguide taper

Apodized grating coupler

Waveguide bend

INVERSE DESIGN
COURSES

Learn from the experts, speed up your research

What you will learn

6Lectures

1 Introduction to Inverse Design In Photonics

13 mins Start

13 mins

In this lecture, we give a broad introduction to the concept of computational design in photonics. We show how simulation and mathematical optimization can be combined to analyze and design photonic devices effectively.

Introduction to Inverse Design in Photonics: Lecture 1 | Flexcompute

View Presentation Slides

2 Inverse Design in Photonics Lecture 2: Adjoint Method

18 mins Start

18 mins

In this lecture, we derive the adjoint variable method for computing the gradient of the figure of merit of a photonic device. We derive this gradient from first principles, starting from the solution of Maxwell’s equations in the frequency domain. Using a simple example of focusing electromagnetic field intensity at a single position, we derive the relevant terms in the gradient and give physical interpretation. We discuss how this method provides gradient information with only two simulations, regardless of the number of design parameters, and how this enables inverse design optimization.

Inverse Design in Photonics Lecture 2: Adjoint Method | Flexcompute

Download .ipynb View Presentation Slides

3 Inverse Design in Photonics Lecture 3: Adjoint Optimization

15 mins Start

15 mins

In this lecture, we show how to use the previously introduced “adjoint variable method” to perform gradient-based optimization of a focusing lens. We set up a simple device based on a pixellated array of permittivity values, compute the gradient of this array with respect to the focusing strength of our lens, and then perform an optimization to achieve a final device that focuses light successfully.

Inverse Design in Photonics Lecture 3: Adjoint Optimization | Flexcompute

Download .ipynb View Presentation Slides

4 Inverse Design in Photonics Lecture 4: Fabrication Constraints

15 mins Start

15 mins

In this lecture, we will discuss the need for fabrication constraints in inverse design optimization. We describe one approach to this and demonstrate it by optimizing a Silicon photonics mode converter.

Inverse Design in Photonics Lecture 4: Fabrication Constraints | Flexcompute

Download .ipynb View Presentation Slides

5 Inverse Design in Photonics Lecture 5: Shape Optimization

15 mins Start

15 mins

In this lecture, we introduce a method to optimize your device with respect to a geometric parameterization using inverse design and the adjoint method. As an example, we demonstrate the inverse design of a 90 degree waveguide bend by shifting the boundaries of the device.

Inverse Design in Photonics Lecture 5: Shape Optimization | Flexcompute

Download .ipynb View Presentation Slides

6 Inverse Design in Photonics Lecture 6: Level Set Parameterization

12 mins Start

12 mins

In this lecture, we introduce a method to optimize your device using a level set parameterization using inverse design and the adjoint method. As an example, we demonstrate the inverse design of a waveguide splitter.