By: Iván Ramírez Morales, Ph. D. Leader of R + D + i in Larvia
@larvia_ai @ivaneduramirez

Up to a few years ago, if we wanted a computer to do something new, we had to program it. That meant coding, step by step, what we wanted the computer to do. Arthur Samuel, in 1956, was determined to create a computer that could beat a human at a board game. So, the question arose, what if we made the computer play with itself as an opponent a million times until it learns winning strategies… and it worked.

In 1962, Arthur’s computer had beaten the Connecticut state champion. Ever since, thousands of new ideas under the same principle have emerged under the name of machine learning or artificial intelligence.

Google’s search engine was undoubtedly the first commercial success of machine learning. Computers can learn to do things that we often do not know how to do ourselves and even do it better than we could have ever done it.

This outcome has led researchers globally to consider taking a step forward, entering images instead of text data and numbers, providing computers with cameras, microphones, environmental sensors, access to internet data, connectivity to other devices, and the ability to move servo motors, with the idea to use this data to train algorithms to perform detection, classification, and decision support tasks.

That approach and the actual demonstrations of its feasibility became a turning point in most industries. Computers can now see, read, hear, speak, move, sense the environment, activate, or deactivate devices, and access data from the Internet.”

Although indeed, there is still no artificial intelligence capable of performing tasks on its own, today’s singletask algorithms are often capable to deliver high performance with a high level of accuracy over long periods, thus far surpassing us in the execution of repetitive tasks.


These developments lead to exciting opportunities in different areas, such as aquaculture. Notably, our company has been capable to detect the location, length, width, weight, and color characteristics of shrimp larvae and juveniles. Interestingly, the mean we used to achieve our goal was a regular app, whose results were more accurate and delivered in less time than an operator could ever provide.

“Our research will soon enable us to estimate relevant characteristics to diagnose health or nutritional problems, just as a preview of the wide list of potential applications.”

The development of cost-effective solutions useful to producers will make an absolute difference so that critical links in the production process can be streamlined and optimized. This defiance generates many opportunities for the new technology startup industry, intending to use artificial intelligence algorithms to solve operational problems efficiently at a low cost.

“We can get ahead of ourselves by assuring that artificial intelligence based applications will replace labor.”

However, the industry has noticed that with technological advances, people are not being replaced by computers, but are integrated into a human-machine symbiosis, generating better productive results in the field, which means that the producer grows, and the job offer increases.

Thus, what would be a challenge turns into an opportunity to optimize production by accessing information that would otherwise be very difficult to obtain.


Ivan Ramirez
Co-founder of Larvia
Research and Development Leader
[email protected]
IG: ivaneduramirez



Please enter your comment!
Please enter your name here