“I believe having a model in a Jupyter notebook, while a common development approach, may limit its usability for others in real-world scenarios. That's where MLOps become crucial, enabling models to be effectively utilized by different users and companies. The growing significance of MLOps lies in facilitating collaboration—making models available to various stakeholders and ensuring continuous monitoring, optimization, and shared access.”
In our new ‘5-minute interview’ series we are going to meet AI and ML professionals who will share their experiences on working in the field. How did they start out and what aspects of AI and ML make them tick? What are their passions and what drives them to create major tech innovations? Follow along on this journey to find out!
In our second interview we talk with David Andres, an ex-aeronautical engineer, who has recently joined the Data Science team at Fever. He is the person behind ‘Machine Learning Pills', a blog that explores the intricacies of Time Series forecasting and Machine Learning.
I am a Data Scientist and currently working at Fever which I recently joined last month. However, I didn't study data or computer science, I actually studied aeronautical engineering and have worked as an aeronautical engineer for around 6 years. Throughout my engineering career, I found myself drawn to programming and working with data. This interest led me to delve into machine learning and deepen my Python knowledge during my free time. The pivotal moment came at the start of the Covid pandemic, prompting me to venture into data science through self-study and a half-year internship. Subsequently, I secured a permanent role as a data scientist, contributing to various projects, including recommendation systems, classification systems, and forecasting models during my 3-year employment in London. Recently, around October, I relocated to Spain and joined Fever.
‘Machine Learning Pills’ began as a documentation of my self-learning journey, aiming to record my progress for personal reference and potential future sharing. Initially, I created brief articles, gradually expanding them as my knowledge deepened. After about 3 to 4 months, I extended this practice to Twitter, inspired by others who shared similar content. The response has been incredible, with a growing community of over 7,500 followers (@daansan_ml).
I believe having a model in a Jupyter notebook, while a common development approach, may limit its usability for others in real-world scenarios. That's where MLOps become crucial, enabling models to be effectively utilized by different users and companies. The growing significance of MLOps lies in facilitating collaboration – making models available to various stakeholders and ensuring continuous monitoring, optimization, and shared access.
Yes. So for example, in my previous role, we were doing our recommendation system, a no-code recommendation system and We needed all these tools to make it available for other people to automatize the training, the data position, everything, every single step, even the monitoring of how well it was behaving so I think that's and a good example.
Well yes, I'd love to mention my own content, 'Machine Learning Pills' but in general, there are numerous valuable resources online. If I had to recommend one person for those eager to learn more about MLOps, it would be Pau Labarta Bajo. His content is exceptional and offers valuable insights for anyone looking to deepen their knowledge in this field.
Want to share your experiences in the AI and ML field with Hopsworks in a future 5-Minute Interview? Feel free to contact us.