I'm a software developer, consultant, and architect with many years of experience in industries of many types and sizes. Seeing how ideas turn into plans and then a has always been a driving force for me. I enjoy taking part in designing and building complex distributed systems, and, in particular, trying to think of ways to combine the power of software with the intelligence of humans to provide the best of both worlds. While being part of CodeValue for the past 6 years, I had the privilege of working with a variety of technologies and platforms together with our customers towards a shared goal of building awesome software.
11:15-12:00 How I built a ML-human hybrid workflow using Computer Vision
While not new at all, Machine Learning has been on the rise of the past years, both because of the ubiquity of data and because of the increase in adoption of Cloud Computing. In recent years, however, ML has become more prevalent than ever - mainly due to its ease of use and its accessibility to non-mathematicians.
In some cases, ML can do things that would’ve been extremely difficult, if not impossible, for us to achieve in the past. In other cases, however, ML is here to assist us, rather than replace us, by relieving us of our most boring and repetitive tasks, and this often has to do with the low accuracy in which ML models operate.
In this talk we are going to build business workflows using the joint effort of humans and software to automate those boring tasks, while compensating for the inaccuracy of ML with human intervention.