As classifying images into categories is a ubiquitous task occurring in various domains, a need for a machine learning pipeline which can accommodate for new categories is easy to justify. In particular, common general requirements are to filter out low-quality (blurred, low contrast etc.) images, and to speed up the learning of new categories if image quality is sufficient. In this blog post, resulting from a joint work with Aigiz Kunafin (Trustbit Data Science), we compare several image classification models from the transfer learning perspective.

DSML projects can be quite different from the software projects: a lot of R&D in a rapidly evolving landscape, working with data, distributions and probabilities instead of code. However, there is one thing in common: iterative development process matters a lot.

In the previous blog post, we created an already pretty powerful image segmentation model in order to detect the shape of truck parking lots on satellite images. However, we will now try to run the code on new hardware and get even better as well as more robust results.

Real-time truck tracking is crucial in logistics: to enable accurate planning and provide reliable estimation of delivery times, operators build detailed profiles of loading stations, providing expected durations of truck loading and unloading, as well as resting times. Yet, how to derive an exact truck status based on mere GPS signals?

Last year I gifted a design systems book I had been reading to a friend and she suggested starting a mini book club so that she’d have some accountability to finish reading the book. I took her up on the offer and so in late spring, our design systems book club was born. But how can you make the meetings fun and engaging even though you're physically separated? Here are a couple of things I learned from running my very first remote book club with my friend!

Process pipeline is a system to break down complex repeatable work into digestable steps that lead to concrete results.Processes, unlike projects, don't have a defined completion time and goal. Their goal is to keep running and moving forward at a steady pace.

I hate distractions. They can easily ruin my day when I'm in the middle of working on a cool project. They do that by overloading my mind, buzzing around inside me, and just making me tired. Even though we can think about several things at once, we can only do one thing at a time.

Black Friday is coming, so let’s talk about warehouse management and event sourcing. When developing a system for event retrieval with aggregates, several very different concepts are possible. If you think of an aggregate as a transaction boundary, then each decision has its own implications.The aggregate can also be a lifecycle boundary - events in a global uniform stream can often only be discarded by the aggregate stem.

Let's talk about celebrating our successes. Our active memory can be like a cache of recently used data. Fresh ideas and frustrations supersede older ones. The older the thought, the fuzzier it becomes. Details are lost.
There is also negativity bias. When good and bad things happen at the same intensity, the bad ones have a greater effect on us.

Like many other knowledge workers, my work day is filled with context switching, hopping from one meeting to the next, and hoping to squeeze in an hour or two of deep work. Having control over my schedule, especially as someone who works part-time, is sometimes just not possible. One of the things I can and do have control over, however, is how I wrap up the work week. Introducing: the weekly shutdown routine!

Knowledge is very powerful. So, finding the right tool to help you gather, structure and access information anywhere and anytime, is rather a necessity than an option. You want to accomplish your tasks better? You want a reliable tool which is easy to use, extendable and adaptable to your personal needs? Today I would like to introduce you to the knowledge management system of my choice: Obsidian.

So far in the series we have built a trivial logistics simulation runtime. At this point it is only capable of finding the fastest route between two locations. This is implemented as a form of A* algorithm that uses predefined travel times.
Let’s extend the implementation and demonstrate how we can “plug” different models into the simulation runtime.

In the previous article (Part 1: Trustbit Logistics Hackathon - Behind the Scenes) we talked about organizing our Logistics Hackathon on Sustainability - how it came to be.
Let’s get to the technical details now.

A look behind the scenes of our Hackathon on Sustainable Logistic Simulation in May 2022. This was a hybrid event, running on-site in Vienna and remotely. Participants from 12 countries developed smart agents to control cargo delivery truck fleets in a simulated Europe.

On Saturday August 7 2021, I released a book called Essential Functional-First F# on LeanPub, a self-publishing platform. It is based on two series of blog posts that I've written for the Trustbit blog.

In this part of composite UIs with micro frontends and design system, we will create a micro frontends that will consume the design system created in the previous post.

As part of their onboarding, all of our employees are encouraged to fill out a “Skwill” — a small matrix that shows an overview of their strengths and interests in different fields. The matrix is “public” (i.e. visible to anyone inside the company) and makes it super easy to share your skillset with teammates or people outside of your team.

On May 13th 2022 the hybrid "Trustbit Hackathon: Sustainable Logistics Simulation" was held at the beautiful Parkhotel Schönbrunn and was also open for remote participants from all over the world. We, a team of three data scientists, took part in it on-site and we want to share a bit about the great experience we had.

We have participated in several hackathons over the years. Accordingly, we have certain expectations for these events. This much can be said in advance, the first Trustbit Hackathon was a positive experience for us. The location in the Parkhotel Schönbrunn was great and the example was well thought out.

At Trustbit, we strongly believe in the benefits of continuous learning. This has not only led to us running an Innovation Incubator but also to developing products that we enjoy using ourselves and are happy to share with other people as well!