Lots of firms sell training on open-source Python tools, so prospective students often ask: How is Coiled Dask training different? How do Coiled’s short, flexible courses help me build the training I need?
The goal of our training sessions is to empower people to solve large-scale data problems through the easiest and simplest means possible, all while leveraging skills they have already. Often, the way to do that is with Dask, Python, and PyData.
Scaling open-source software is not always as easy as demos and tutorials make it look. We aim to help people use Dask to scale OSS technologies that are widely known, used, and loved, from Pandas to PyTorch. In training participants to build intuition and understanding around distributed computing, we enable them to use unique tools like the real-time Dask dashboards to gain insight for real-world problem-solving.
Training designed to help real people solve real problems
We provide deep technical expertise that aligns with the problems that real people actually run into. Instructors showcase solving those problems across multiple tools and platforms — including Dask, XGBoost, Tensorflow, and more … and we won’t tell you that Dask/Python is always the answer (because it’s not!). We’re prepared to deep dive into your use cases as part of the class and we won’t rely on oversimplified assumptions. If you bring us a specific example from your work, we’ll get specific with you in solving it.
Structured to fit your needs and your schedule
When building out these training courses, we were user-focused and problem-focused, rather than product-focused. We structured our offerings around the real-world missions users have, such as transforming data or understanding parallelization. By design, this is not a one-size-fits-all platform boot camp (though we’re not opposed to running a boot camp down the line if people want a full platform discovery experience).
To make training easily customizable, we keep our courses small and granular. Lots of attendees want to use Dask DataFrames and we want to deliver. At the same time, since many business data users work with traditional tabular data and will never use raw multidimensional array data, there’s no reason to ask those participants to spend hours with an unfamiliar array paradigm — even though the array data processing model is a key goal for others. Meanwhile, still other users want Dask to parallelize their custom algorithms and don’t want to learn about another dataframe API.
With this in mind, we offer flexible course lengths to accommodate different needs and backgrounds. As another example, we know that some Dask users are new to machine learning, while others are veterans of many frameworks. As a result, everything is easily customizable since we understand that time and attention are at a premium.
What’s more, we promise our classes aren’t just good excuses to deliver infomercials for Coiled Cloud. We do think that Coiled Cloud is the most efficient and effective way to manage Python at scale for many organizations, but we don’t tilt the table in that direction, and we even discuss free/OSS alternatives for several Coiled features.
Now is the time to take your data science and ML to the next level
We know that all distributed computing has some inescapable complexity. We believe, however, that Dask+Python offers by far the simplest and most intuitive toolset for today’s problems. We’re excited to learn with you. Click below to learn more about our upcoming training sessions and to secure your spot.