Your AI in the sky
Next generation weather intelligence
AI powered hurricane intensity forecast
World’s lightest weather sondes
WHAT CAN WE DO FOR YOU?
Working with us, weather measurement institutions, weather stations and atmospheric researchers can measure weather more precisely while reducing the burden of weather measurement debris into the environment. Additionally, insurance companies, financial institutions and emergency management professionals are able to provide services for people and businesses with our AI-empowered severe tropical storm forecasting model.
Skyfora combines scientific understanding of tropical storms with computer vision, machine learning and IoT technologies. Skyfora’s power pair of weather intelligence consists of the AI empowered Tropical Storm Tracker and lightweight weather instruments StreamSondes.
Tropical Storm Forecasting
- Forecast update every 7-30 min
- Resolution of 250 m – 1 km
- Best forecast accuracy 0-24 h all over the world
- Intuitive user interface or tailored API
StreamSonde / Weather Instruments
- Worlds lightest weather sonde
- Data collection during ascent and descent
- Low weight and slow velocity make parachutes redundant
- Increased and Improved wind data
Copernicus Masters Winner
Awarded in the highly respected Copernicus Masters and Galileo Masters innovation competitions. Part of European Space Agency’s Business Incubator Centre (ESA BIC) until end of 2020.
TROPICAL STORM TRACKER
The Tropical Storm Tracker software provides the most up-to-date weather data possible. A machine learning model issues storm intensity estimates and forecasts with every incoming satellite image, numerical weather forecast and dropsonde measurement. Application segments include insurance linked security funds (ILS), (re)insurance, emergency and disaster management, aviation and shipping.
The StreamSonde is the world´s lightest weather sonde. It weighs less than 15 grams, and it is used as a radiosonde, a dualsonde (i.e. radiosonde during ascent and free falling dropsonde during descent) or as a dropsonde. In all user scenarios, it provides numerous benefits.
“Weather provides some of the most demanding tasks for data processing and forecasting. There is a need to acquire and process weather data faster, in more detail and more reliably than ever before.”
Computer Vision for situational awareness
Today, our CEO Svante Henriksson is speaking at this interesting seminar about Computer Vision for situational awareness in the seminar arranged by Silo AI.
Our computer-vision-based forecast product is featured in today’s Tekniikka&Talous
Our computer-vision-based forecast product is featured in today’s Tekniikka&Talous.
Hurricane Unwinder at ESA business incubation centre
Hurricane Unwinder at ESA business incubation centre
The Skyfora team features a unique combination of expertise in weather instrument engineering, atmospheric
science, machine learning and international sales.
(PhD) Founder, CPO, Climate Scientist
(PhD) ML Engineer
CTO & Board
Senior Scientist, Advisor
Embedded SW Engineer
We invite you to apply and join us on an exciting pioneering journey in building something brand new with significant impact on the real world.
Skyfora develops next-generation weather technology. The StreamSonde is world’s lightest weather sonde and the AI-powered weather forecasts bring solutions to many industries in a world faced with climate change and extreme weather.
Skyfora’s high-tech solutions increase the amount of weather and climate change observations and turns that data into better weather and climate forecasts and risk data through artificial intelligence backed up by academic research. We have won prestigious innovation competitions and our products are sold internationally.
The company has several patents, which fuels the company’s giant leap in atmospheric data gathering and improved weather forecasting. The team features a unique combination of expertise in atmospheric science, weather instrument engineering, machine learning and business development.
See open vacancies below:
Machine Learning Engineer
Your main responsibility will be in applying deep learning on forecasting of various weather phenomena from hurricanes to hyperlocal weather forecasts within a city.
Most of our forecast products combine multiple data streams from various sources, including numerical weather prediction, openly available measurement data and data from Skyfora’s proprietary instruments. Deep learning technologies used at Skyfora today include CNN, time series models, RNN, BNN (Bayesian Neural Network) and multimodal architectures.
The backend of our service runs on AWS. Some of the tasks beyond developing the algorithm are managing the pipeline of the large volumes of weather data stream from original sources to cloud storage to output on our graphical user interface.
Experience & Skills
We hope that your experience includes:
- Deep learning, CNN/RNN
- PyTorch or Tensorflow
- Working with and building complex datasets
- Knowledge of cloud services such as AWS, GCP or Azure; ML frameworks such as SageMaker is a plus.
Related work experience is naturally a plus and will be reflected in the compensation.
An interest in the topic and the natural science in general is necessary for you to have the right motivation, but no pre-skills are necessary as these can be learnt from the team.
Compensation & benefits
We offer a flexible and pleasant work environment with a low-hierarchy team with possibility for remote work. The compensation is a mix of a competitive monthly salary, fringe benefits and possibility to get stock options depending on your level of expertise.
We invite you to join us on an exciting pioneering journey in building something brand new with significant impact on the real world. You’re more than welcome to get in touch for further information.
Send your application to, or feel free to ask for more information from svante.henriksson(at)skyfora.com or phone number +358 50 408 7900.