Decorative image frame


First-Hand Insights on ESG-Friendly AIoT(AI+IoT) InsurTech


Coronavirus Italy - How to analyse like a data scientist

Coronavirus Italy - why we should be concerned when we predicted accurately

We published our predictions on the Coronavirus confirmed cases for 15th March in our previous post Coronavirus Italy statistics and predictions. Our predictions were 24734~25422, the actual number was 24747, the errors of prediction were -0.05% ~ 2.7%.

However it was not something worth of celebrating at this moment in this particular case. On one hand, the accurate preditions indicate that the development of Coronavirus in Italy had a pattern and we were good at finding this pattern. On the other hand, it also meant that, up to yesterday the 15th March, the Coronavirus situation in Italy were still following its historical pattern, which could lead to a bad closure of this pandemic outbreak.

If, in the next days, we don’t see abrupt down slopes in the day-to-day incremental curve, we might be looking at a “Gradually Contained Scenario” with a total confirmed infected population of 80,000+, overtaking China and becoming “number 1” before the end of March.

It’s not what we would wish. In this specific case, we’d rather be wrong at predictions and we look forward to seeing concrete, even better drastic, signals of improvement from numbers.


Coronavirus Italy statistics and predictions

Coronavirus Italy - before it gets better, it might get a lot worse

It’s been slightly more than 3 weeks since the 20th February, when the “Patient 1”, a 38-year-old man, checked himself into a local hospital in Codogno, a small town in Province Lodi, Lombardy Region. The present “Patient 1” offically unveiled the outbreak and, in less than 1 month, Italy has become to having the highest number of confirmed positive cases and deaths outside of China, with 1441 deaths and in total 21157 people infected throughout all 20 regions of the country. Up to today, the 14th March 2020, the spread of the infection has not yet shown a consistent signal of slowing down in spite of the national-wide lockdown announced on the 8th March.

On the 27th February, Italy has already decided to put 11 towns on lockdown to contain the virus’ spread. From the following statistics, it is obvious that the virus has long transcended the boundaries of these 11 towns and has found its way throughout the country and overseas.

It’s still hard to predict when it will evetually reach the tipping point. We have conducted some exercises to predict the trend in the next days:

  • Time of the prediction 2020-03-12
    • 2020-03-13: 17380 (actual: 17660, error -1.6%)
  • Time of the prediction 2020-03-13
    • 2020-03-14: 20309 (linear) - 20584 (log) (actual: 21157, error -4% to -2,7%)
  • Time of the prediction 2020-03-14
    • 2020-03-15: 25422 (linear) - 24734 (log)

As TECHENGINES, we have prepared open datasets with up-to-date official statistics on GitHub: coronavirus-stats-italy together with up-to-date prediction trackings.

Coronavirus Italy - the next 5 days are critical from China experience


Coronavirus Italy open datesets updated by 6th March 2020

Open datasets on GitHub to help analytics on Coronavirus outbreak in Italy

What’re the latest stats of coronavirus in Italy? How fast it has been spreading? What would be the trend? We would like to do our little contribution for better understanding the 2020 coronavirus outbreak in Italy. Check out TECHENGINES open datasets with up-to-date official statistics on GitHub: coronavirus-stats-italy

For benchmarking purpose, we also prepared the datasets with the historic stats of coronavirus spread in China. We believe that good data is the foundation of valid analysis.



Why On-Demand Insurance? Why InsurEngines?

On-demand Insurance is revolutionizing the relationship between insurers and consumers. Insurers need to rethink their strategies, products and business models to stay relevant in such a fast-evolving market.

TECHENGINES InsurTech AI solutions, InsurEngines, help to accelerate the implementation of innovative On-Demand Insurance, thanks to its best-in-class backbone systems, proven analytical excellency and its capability to extract effective value from the top trending advanced technologies.

Traditional insurance companies have adopted policy-centric approach that, although it profited insurers for decades, has led customers to think about insurance typically only in two occasions:

  • To buy/renew a policy
  • To make a claim

Insurers need to switch from a policy-centric approach to a customer-centric approach in order to exploit the huge undiscovered opportunities from close interactions with their customers. As in the digital age, long period without contact or being perceived value-added, becomes an easy entry point for other new and innovative competitors.


TECHENGINES AI FORCE - Free Course Data Science 101


TECHENGINES constantly looks for like-minded collaborators who are passionate with learning the new AI technologies, discovering the hidden value inside the data and fostering the beneficial use of data science for a better world. Even if you are new beginners, as long as you’re mentally and academically prepared, we can help you to acquire the skills and experience needed to become a data scientist.

In the last few years, TECHENGINES has developed a series of AI solutions in the insurance and credit sectors that are verified by well-known international organizations. Thank to those valuable experience, we are able to design the training courses that are practical and related to concrete real-life objectives.

TECHENGINES’ solutions are developed in various programming langurages such as Python, Scala, Javascript and C++. However, our training courses are mainly in Python. Python is the most widely used programming language in the Data Science field. The training course Data Science 101 will provide the beginners with a chance to dwell into the Python world and explore its fast end effective functionalities in processing the data.


AIoT INSURANCE - The ultimate key to the future insurance

AI can effectively interprete complex IoT data so to empower AIOT Insurance

IoT (Internet of Things) has long gone beyond a buzzword and its surge is unstoppable. Every second, countless sensors are generating terabytes of real-time data points. IoT enables the network of “things” to observe, record, understand and guide situations without human presence. IoT has to affiliate with AI because only machines can tackle with such a tremendous amount of data. Many new opportunities and markets have begun to appear with the convergence of AI and IoT, according to IDC (International Data Corporation), insurance is the second industry that’s most impacted.

Insurance applies actuarial science that uses mathematical and statistical methods to analyze historical portfolio and loss data, to assess present risks and to predict the future events. In the era of big data, insurance has to take more proactive control of risk selection and risk prevention. AIoT, the combination of AI and IoT, is the ultimate key.



New Year’s day, after the breakfast, I am scrolling through as usual my list of freshly published articles and papers on the most recent Machine-Learning development in order to stay on the frontline of the AI evolution. Among all the inspiring experiments, one article has caught my eyes: With 99 lines of code, you can have the magic of snow and ice like Princess Aish in Frozen. It’s an article about a new programming languageTaichi designed to provide native support for sparse data structures, such as graphics.

In this very first Edition of our AI Frontline column, we’d like to present you in more detail this productive programming language Taichi and share with you our views on the Top AI Trends in 2020.




TECHENGINES is an Innovative Tech Startup that provides AI Engines and Algorithms to accelerate innovation for InsurTech | FinTech | IoT Analytics. Our Mission is to improve insurance and credit risk selection and help to boost industry efficiency by staying at the frontline of AI evolution and offering the state-of-the-art AI solutions.

With the start of a new decade on the horizon, it’s time to take a step back and reflect the work we’ve done in order to better look forward and prepare to welcome 2020.

In 2019, we made progresses in developing a broad set of innovative solutions for/with our Clients and Partners that have brought concrete added values to their business.

We’re very happy that in 2019 we were able to win the Italian National Grant Smart&Start as an award for our efforts in pursuit of our Mission.

Here below is a list of our main solutions: