Databricks, Qlik, and Nvidia show AI is heating up
We knew AI was hot, but last week brought several announcements that show the industry is trying to bring AI and machine learning (ML) into the mainstream of business. With two AI acquisitions by big names in the analytics space and a $ 100 million funding round for an MLOps specialist, it has been a week of AI acceleration.
Brick and bamboo
Today’s news is that Databricks is acquiring German company 8080 Labs, maker of bamboulib, a user interface-based data science tool that generates code, but doesn’t require users to write it. And that follows Databricks’ announcement earlier this year that it had added AutoML functionality to its “lacehouse” data platform.
Also Read: Databricks Boosts AI With New AutoML Engine and Feature Store
Clemens Mewald, Director of Product Management, Data Science and Machine Learning at Databricks, spoke to ZDNet and explained that bamoboolib is currently offered as a Jupyter laptop plug-in and will be , logically, integrated into Databricks own laptops. Subsequently, it will also be integrated into the user interface of the Databricks workspace, to make it accessible to less technical users, often called “citizen data scientists”.
Mewald explained that the acquisition aligns with Databricks’ approach to expose several layers of abstraction: at the coding level for data scientists, machine learning engineers and, of course, data engineers; at the SQL level for professionals more focused on databases as well as for professional users; and at the user interface level for less technical users who are still passionate about their data and their analysis.
AI + BI just … Qliks
While it comes down to integrating AI into data analytics platforms, that’s what Qlik had in mind when it announced the Big Squid acquisition on September 30. Big Squid offers an AutoML platform that will continue to be offered as a standalone Qlik AutoML, but will eventually be integrated with Qlik Sense. This will allow business intelligence (BI) users to use their datasets for training models, and then evaluate additional data against the models, with predictions brought back as new columns in the dataset. , where they can be easily viewed like any other data.
Big Squid’s technology even provides the explainability of the ML model, through the use of Shapley values, which Databricks AutoML also does. The fact that both vendors provide the explainability of AI based on Shapely’s value is another sign of the AI company’s ambitions, as some data protection regulations require models to IA, in fact for a matter of trust and auditability.
Qlik’s goal, like Databricks’ goal, is to make AI accessible to analytics teams, not just data science teams. Specifically, Qlik focuses on key factor analysis, predictive analytics, and “what if” decision planning, all integrated into the BI environment and paradigm. And, yes, Qlik wants these “citizen data scientists” too. Additionally, armed with technology derived from its many other acquisitions, such as Podium Data (now Qlik Catalog) and Attunity (now Qlik Data Integration), Qlik’s presence throughout the data lifecycle and its governance, gives the company a good MLOps (machine learning operations) also play.
Read also :
More money … and GPU integration
Speaking of MLOps, this is the very area Domino Data Lab has focused on, and more specifically in the enterprise context. And just yesterday, Domino announced its $ 100 million Series F round of funding, led by Great Hill Partners, with existing investors Coatue Management, Highland Capital Partners and Sequoia Capital and new investor Nvidia. Nvidia, of course, is the leading vendor of GPU chips and servers, all of which can speed up AI training and scoring, especially in the area of deep learning.
Read also :
In fact, Domino and Nvidia used yesterday’s funding announcement to reveal their own strengthened partnership. According to its press release, the partnership will involve Domino’s development of product features to extend the accelerated computing capabilities of its platform, including “validating the Domino platform for NVIDIA AI Enterprise so that Domino can operate from seamlessly on consumer systems certified by NVIDIA from OEM hardware vendors. ” Domino also said the two companies will market these products together.
What has AI been doing for me lately?
The fact that these three deals were announced within a week shows just how important AI / ML is in the context of enterprise IT. It is possible, if not likely, that this first wave of AutoML integration into BI and analytics platforms is just a first iteration of the IA + BI equation. The automation of ML algorithm selection, hyperparameter optimization, feature selection, and even explainability is great. But eventually, it will also be necessary to automate feature engineering, model deployment, monitoring and recycling.
Also Read: DotData Offers Automated Feature Engineering For Databricks
Things are happening fast enough that these first generation innovations can become proverbial “table stakes” very soon. As analytics companies differentiate themselves from the competition, the world of AI can become more interesting, relevant, accessible, and hopefully more easily scrutinized and trustworthy.