All blog posts

Veracell in 2022 - projects, tools and technologies

We looked back to summarise year 2022, and wow it is a lot! We worked in projects of very different types and sizes, from years-long continuous development to few days of audits and strategies. Mostly, we did data science and data engineering to deliver AI and cloud solutions, spiced up with a bit of front-end and visualisation.

Most projects started with workshops for service design and data strategies to pin down use cases and needs for technical solutions. After setting clear goals, we planned data architecture and created data analytics.

All in all, in 2022 we delivered the whole chain from proof of concepts, pilot deliveries, migration projects, and production deployments to continuous deliveries.

Our most important projects included a platform for a market research tool, massive healthcare cloud projects, healthcare services from IoT-enabled monitoring devices, and machine learning models to predict when spare parts need to be changed. We worked with plenty of things from C algorithms for microcontrollers to data analytics for PowerBI reports.

For development, we mostly used Python programming language, TypeScript and React, with a hint of embedded C. Our projects are cloud native - we worked in AWS, Azure and GCP.

For documentation and project management, most used were GitLab, Azure DevOps, Confluence and Jira.

PostgreSQL was most popular for data storage and GraphQL as an API for the frontend. Cypress and Playwright were used for end-to-end testing and Chart.js, Plotly.js, and PowerBI for data visualisation.

For AI, we implemented Random Survival Forest (RSF) model with AWS Sagemaker and AWS Athena. Natural language processing (NLP) is trending - we developed and are developing NLP solutions for marketing research and healthcare applications. We also used Bayesian optimization, gradient-free optimization, random&grid search as well as reinforcement learning.

To dig into details, we created messaging using MQTT protocol and Kalman filters for positioning. Among other unmentioned tools and technologies, we used at least Miniconda, Pandas Dataframes, VS Code and Spyder IDEs, Linux, GNU Screen, PHP, HTML and CSS.

In short, we set up development environments, integrated data to centralised platforms, modeled data structures, processed data for analysis, designed and developed APIs and DevOps services, and delivered everything in a neat package of documentation or visualisation.

Last but not least, we asked about our culture and the employees answered.

At Veracell, we value good atmosphere, salary, and flexibility. We use interesting technologies to deliver impactful projects. Our people are young, nice, ambitious and honest and as a company we are small and forward-looking, yet still searching for future direction. This sets us for an interesting path for the year 2023.

Curious to know more? Check out our references below.

Do you want to find out how to harness data to grow your business? If so, we would love to have a chat.