One in all the things that I each love and hate is the chance to establish a recent data stack. Every time I feel, “this time I’m going to get it right”. And each time without fail, I’m stuck coping with the tradeoffs of all the choices of the “modern data stack”. Simply put:

• it’s too complicated;
• it takes too long;
• it’s too expensive; and
• talent is tough to get.

I still remember those early days of the COVID response once we were working around the clock to gather data to construct and evaluate models, answer basic questions for policy makers, and understand what was going down in hospitals. We would have liked a contemporary data stack fast — I mean inside 24 hours fast. We didn’t have time to assemble requirements or work out what technologies play nice with one another. I needed a contemporary data stack with the push of a button where I could select different technologies and deploy them without weeks of configuration — identical to I can do with iOS or Android and have the apps vetted from their respective stores.

Fast forward to a couple of months ago once I was lucky enough to fulfill Yevgeniy, Stella, and Christine. They knew my pain about complexity, fragmentation, deployment and compatibility challenges of all of the technologies available to data engineers and data scientists. And that’s why they began Shakudo — to construct the operating system for the trendy data stack.

Shakudo is the fastest/easiest option to construct, deploy and manage a best-in-class enterprise caliber data stack. It unites essentially the most widely used data, AI/ML tools and scaling frameworks like DBT, Airflow, Superset, Spark, Ray, Dask and up to date tooling in generative AI resembling LangChain, ChromaDB, Milvus and leading LLMs right into a single platform! And it does so with scalability, compatibility and robustness without a military of information engineers.

They supply the pliability to decide on the best data tools while solving the compatibility challenges that every one too often arise when different teams inside a corporation attempt to create their very own data stack (e.g., marketing and sales constructing one custom data stack vs the information stack for the core services). The great thing about Shakudo lies within the incontrovertible fact that a knowledge stack will not be a static piece of technology encased in glass but quite a living and evolving system that might be modified or added based on an enterprise’s requirement.

I’m pleased to announce that GPV is leading the Series A round for Shakudo, with participation from existing investors and I’ll be joining the board. Shakudo is central to our thesis at GPV for a way the information world goes to evolve. And particularly to be literally the inspiration for deploying AI solutions. With their wealthy mix of leadership roles, technical skills, and industry experience, Yevgeniy Vahlis, Stella Wu, and Christine Yuen are the best team to tackle this problem. Their collective expertise in bringing artificial intelligence and data use cases to large scale production through their work in big tech, banking, consulting and growth stage VC signifies a deep comprehension of how advanced technologies might be seamlessly integrated to create efficient and effective data solutions.

Don’t take my word for it. Go try it for yourself!

And we’re growing so join us on this exciting journey!