Snowflake Data World Tour 2024 in Sydney
Last week, the Snowflake Data World Tour 2024 made its exciting return to Sydney, bringing together data innovators and industry leaders to share the latest in data technology. We were thrilled to be part of the event, supporting our friends and event sponsors at KADA – a cutting-edge data governance platform that’s transforming the way organisations manage and utilise their data.
Curious about KADA? KADA empowers businesses to gain better control over their data by preserving critical knowledge, preventing duplication, and ensuring data assets are continuously leveraged for new insights. With KADA’s technology, organisations can unlock unparalleled visibility into their data, accelerating the path to smarter, data-driven decisions. Interested in enhancing your data governance? Learn more at https://www.kada.ai/
Key highlights from the conference
1. Document AI – Pioneering Unstructured Data Processing
The Snowflake team showcased their advanced Document AI, designed to revolutionise unstructured data processing. Here’s what stood out:
Advanced Data Extraction:
Document AI allows users to choose from some of the most sophisticated open source AI models such as Mistral and Llama, as well as snowflake’s very own Arctic-TILT (Text Image Layout Transformer). Additionally, Document AI is multimodal meaning it can process various types of inputs such as text, images, tables without requiring any prior knowledge. This opens up the potential for a vast amount of unstructured data to be processed within Snowflake.
Streamlined Workflow:
Snowflake has enhanced the development process for Document AI, enabling users to define, train, evaluate, and deploy—all within Snowflake itself. Not only can developers stay completely within Snowflake to develop a model but quickly too! Users provide a prompt to the model in natural language then the document is parsed and returned in a structured format. To further streamline operations, Snowflake allows Document AI to run as a task allowing processing of future documents to be automated.
2. Time-Series Analytics in Snowflake
For those working with time-based data, Snowflake introduced advanced time-series analytics functions designed to simplify complex time-data manipulations. Key functions include:
TIME_SLICE – Helps partition data into time-based buckets for better analysis.
ASOF JOIN – Allows users to join time series datasets even when timestamps don’t exactly match, bridging gaps in datasets.
RANGE BETWEEN – Creates moving windows for analyses that aren’t impacted by data gaps, ideal for trend analysis and forecasting.
Moreover, users can build custom functions with Snowflake’s existing tools, such as ASOF JOIN for interpolating and filling in missing values. These functions are game-changers for those working with IoT or sensor data, providing more refined tools for predictive analysis and trend forecasting.
Celebrating Data-Driven Impact: A Look Back
As we reflect on the conference’s highlights, one of our proudest moments was the recognition our client CES received last year with the Snowflake ‘Data for Good’ Award. This honour celebrates CES’s dedication to making a difference through data and highlights the collective impact of our collaboration with CES and KADA.
We hope you enjoyed our team’s reflections on the Snowflake conference and found a few useful insights along the way.
If you’re looking to make the most of Snowflake’s capabilities, do get in touch!
Check out our vlog of the day:
This blog was written by Jack, Natasha and Raveena.
About EdgeRed
EdgeRed is an Australian boutique consultancy specialising in data and analytics. We draw value and insights through data science and artificial intelligence to help companies make faster and smarter decisions.
Subscribe to our newsletter to receive our latest data analysis and reports directly to your inbox.
