Chasing Predictability with AI: The Model of You Outperforms You
Usually, the first question a client asks about software development is: "When will it be done?". Traditional methods to answering this question are fraught with errors. The most common errors include heavy reliance on estimates and the use of averages to give a deterministic answer. What goes through your mind when you try to answer this question? Now imagine that we could take the same process that your mind goes through and model it? Get rid of the biases that we as humans have by using the data our systems already track. Advances in AI, combined with the rapid growth of data across industries, has prompted the exploration of novel methods for data analysis and forecasting. Join Prateek Singh as he talks about techniques that are over 100 years old, but apply the same principles we use today. With very little estimation and simple measurements, predictive models can better inform and equip teams, managers, and customers with information about possible completion dates of software projects. You'll discover exactly what these techniques are and how you can use your data to do the same thing the National Hurricane Center, sports betting websites, and financial advisors do—make better predictions than we do!