내가 처음 이 회사를 알게 된 건 2019년 여름이다. GDG Seoul 모두의 Toy Story 라는 행사에서 지인이 발표한다고 하여 들으러 갔다가 한 발표가 굉장히 흥미로웠다. 한창 머신러닝에 관심을 가지던 시기였는데 머신러닝에 활용하는 GPU 가상화 관리 도구에 관한 내용이었다. 굉장히 기술적으로 깊이 있고 이런 걸 하는 회사도 있구나 하고 처음으로 래블업에 대해 알게 되었다.
In this article, we introduce FastTrack, the MLOps platform of Backend.AI. With FastTrack, users can compose each step of data preprocessing, training, validation, deployment, and inference into a single pipeline. Especially, FastTrack allows users to easily customize each step when building a pipeline. In this post, we introduce the background of Backend.AI FastTrack and its unique features, along with why MLOps platforms are needed.
As the complexity of a program increases, software developers need good debugging tools. The ideal debugging process is to identify how to reliably reproduce a problem in a development environment where you can experiment freely, and then create automated tests based on this. However, when the reproduction scenario itself is too complex or the kind of bugs that only occur randomly in production environments, the next best thing is to log in detail so that you can understand what the problem was later. In this post, we introduce the aiomonitor-ng tool we developed to make debugging complex asyncio programs easier.