Demetrios is the founder of the MLOps Community. He likes interviewing people and playing improvised songs on the guitar.
The MLOps Community would not be possible if it weren’t for the amazing sponsors we have. Here is a brief update on what they are doing these days. Double-click on any of them to learn more.
Redis is bringing real-time ML to the masses. It’s flexible data structure and real-time capability caters well to the data lifecycle for AI/ML. It’s been widely adopted for online feature stores to supplement productionized AI/ML use cases, and as a vector db for semantic search.
2022 was interesting. We raised an $8m Series A, and got to the top 100 AI startups list by CB Insights. We rolled out way better support for model registry use cases and revamped our docs and website.
Most importantly, we stayed true to our vision of building a reliable component for model metadata management that you can plug in, and it “just works”. In 2023 we’ll continue doing just that, improving developer experience around experiment tracking and model registry for production ML teams.
Look out for a new product debut from Petuum in 2023! Their AI automation platform combines the groundbreaking capabilities of LLMs with the rapid time to value of workflow automation. Sign up for the wait list at www.petuum.com!
Flyte’s v1.3.0 release in early 2023 will have feature signaling – the ability to pass data to a running workflow, driven by human interaction. Till then, take Flyte for a test drive using the Union-hosted Flyte Sandbox at sandbox.union.ai
Wallaroo’s Free Community Edition helps data scientists and ML engineers deploy models in seconds through UI, SDK, and self-service APIs, and experiment, test and iterate models with confidence thanks to comprehensive model observability.
Arize AI had quite a year, launching new tools in its self-serve ML observability platform to tackle AI fairness, automate AI ROI calculations, and monitor embedding drift for CV and NLP models (see their lightning talk on the subject).
Snorkel AI will be holding a virtual Foundation Model Summit on January 17. Come learn about enterprise applications, best practices, and the latest research in the field of foundation models (FMs) and large language models (LLMs).
Run.ai has open-sourced some key tools to control, configure and monitor the GPU resources that you are using. In 2023 they are continuing efforts to make the life of AI/ML teams easier, so keep an eye out for them. Thanks to all MLOps Community members and have an awesome GPU accelerated 2023!
In 2022, Tecton expanded on its vision to build the most complete feature platform for real-time ML, helping ML teams solve many of the common challenges of getting ML to production. We’re looking ahead to the new year and many more exciting projects!
2022 has, without a doubt, been all about scale. Our highlights? On the top of our list, our community edition hit +1,000 users, we rolled out private cloud deployments for all the major clouds to accelerate enterprise adoption, and Superwise Projects was a huge step forward in streamlining high-scale ML observability management. See you in 2023 with some super interesting releases for the community ❤️ Superwise
2022 was an astronomical year for Galileo 💫
We launched our Machine Learning Data Quality Engine and helped users improve model accuracy by 10-15% across the board along with 30% savings in labeling/data acquisition costs! We also launched our free community offering of Galileo– now anyone can instantly debug their ML data with only a single line of code & get model performance wins instantly 🚀
2022 was the path-breaking year for advanced AI models. To cater to this explosive growth in model architectures, Fiddler rolled out ML Monitoring and Explainability capabilities for unstructured data like text and images and transformer models trained on these datasets.
Check it all out here.
We are currently looking for 2023 partners. If you are interested in our sponsorship packages feel free to reach out.