PinnedInTDS ArchivebyBenjamin Thürer·Nov 27, 2023Pre-Training Context is All You NeedThe driving force behind modern transformer models stems to a large extent from its pertaining data, allowing for strong in-context…A response icon5A response icon5
PinnedInTDS ArchivebyBenjamin Thürer·Nov 8, 2023The Inflation of AI: Is More Always Better?New ML models emerge hourly but this fast pace comes with drawbacks; hypothesis-driven development can help to mitigate those.A response icon7A response icon7
PinnedInTDS ArchivebyBenjamin Thürer·Oct 10, 2023Don’t Forget Confidence Intervals for Your ML ProductMachine Learning is never 100% correct. Thus, an ML model is only helpful when users understand the uncertainty of predictions.A response icon5A response icon5
InTDS ArchivebyBenjamin Thürer·Jan 21, 2024How to Low-Pass Filter in Google BigQueryWhen working with time-series data, it can be important to apply filtering to remove noise. This story shows how to implement a low-pass…A response icon4A response icon4
InTDS ArchivebyBenjamin Thürer·Oct 25, 2023How to Avoid Five Common Mistakes in Google BigQuery / SQLWhile working with BigQuery for years, I observed 5 issues that are commonly made, even by experienced Data Scientists
InTDS ArchivebyBenjamin Thürer·Jun 20, 2023Rotating On-Call for Operational and Support: A Must for Data TeamsA rotating on-call schedule for operational, support, and tech-dept frees the rest of the team to do great development
InTDS ArchivebyBenjamin Thürer·Jun 7, 2023Daily Standups May Not Improve Your Team’s AgilityIf you feel daily standups are not helpful for your data science team, you might be rightA response icon4A response icon4
InTDS ArchivebyBenjamin Thürer·May 30, 2023Success in ML Projects through Technical DrawingsImprove workflows and expectation management in ML through technical drawingsA response icon1A response icon1
InTDS ArchivebyBenjamin Thürer·Mar 9, 2023How Machine Learning Will Become a Game Changer for the Location Data IndustryLocation data can provide unique insights but comes with costs and privacy issues. ML can overcome these drawbacks and improve location…A response icon1A response icon1