Snowflake Summit Stories: A Tech and AI Pilgrimage

---

I shall have to begin with a personal narrative before moving on to the topic of technology and innovation, I promise.

To me, Snowflake is far more than just a cloud-based data platform. It's much more. The community and the people who surround it are what make it unique. It was both a blessing and a curse for me to be highly involved in the platform adoption in my previous job. Inevitably, you only recall the good things after some time has passed. I have encountered super bright Snowflake employees, ranging from executives to professional services and solution engineers. They became not just my buddies but also, in a sense, my mentors. The nicest and most knowledgeable people in the tech industry. They have helped me learn and acquire knowledge that I thought was out of my reach. I am forever grateful for allowing me to grow because, in my experience, not many companies and providers do this. (I am not even a Snowflake employee.) And when I grow up, I want to be Alistair Hankin.

We were blessed with professional lighting during the Amsterdam Summit

The Snowflake World Tour is a multifaceted "Nerdfest" where you get to know the complete technical team as well as regional partners from consulting firms and other software companies. Numerous tracks with customer stories, engineering solutions, and hands-on training. Russian roulette is my only method of selecting because it's usually very hard to choose between the talks.

Seeing my old friends from the Salesforce community, SDG Group, Dataiku, dbt Labs, and EPAM Systems made me super content both in Amsterdam and Berlin. It was enjoyable to spend a day engaging with like-minded people; I made some new friends from the pharmaceutical industry and met some former coworkers. I don't mind the Zaandam location (for the Amsterdam summit) because it was close to my home, but the summit in Berlin had a more pleasant atmosphere in the city center. Overall, it's always incredibly well-organized, and when you love your stay, time flies. And the Ikea concept always works, providing food and drinks to the attendees, and they will stay happily with a full belly.

I only noticed once through these two events that the speaker went over his notes and PowerPoint. It's disappointing; you immediately lose attention. The majority of the sessions were crowded, and I always left feeling a tad smarter. Snowflake made the presentation easier to digest for a broad audience by taking into account that attendees are not always die-hard tech dudes and dudettes.

Data mesh was the buzzword a few years ago. It's clearly AI now, and it's glued to all of the main topics and panel discussions like the label to the pickle jar.

Since data mesh first appeared on the surface of the Earth, we have been talking about data products, which might mean different things to each of us, yet we keep using the term.

Other than the data product, domain-oriented data ownership, and a self-service platform are part of our current data universe as well. Creating a self-service capability is the ultimate goal of implementing a cloud data platform such as Snowflake, but not all businesses are achieving it.

In Daiichi Sankyo's presentation, the Sr. Data Platform Engineer finished his speech by saying that data mesh relies on the culture, and it's not a universal one-size-fits-all solution. I couldn't agree more with him.

There is an excellent presentation by Roche on their data mesh journey if interested. https://www.youtube.com/watch?v=umCZJKMqLpY

Cortex AI is Snowflake's cerebral cortex. Yes, I've always been interested in playing with the word. Your cerebral cortex is the largest location of neuronal integration in the central nervous system, and it is crucial for attention, perception, awareness, thought, memory, language, and consciousness. Allow me to indulge my nerdiness. Fun fact that a man's prefrontal cortex, which controls executive processes like planning and decision-making, keeps growing until his mid-to-late 20s.

In a nutshell, Snowflake's Cortex is a set of AI tools that employ large language models (LLMs) to understand unstructured data, answer freeform queries, and provide intelligent support. Then there's Snowflake ML, which gives you the ability to create your own models. Cortex AISQL in Snowflake piqued my interest the most. Which uses industry-leading LLMs from OpenAI, Anthropic, Meta, Mistral AI, and DeepSeek to do unstructured analytics on text and images. For structured data, Snowflake offers a useful feature called Cortex Analyst. Using your structured data in Snowflake, this LLM-powered functionality enables you to develop applications that can consistently respond to business inquiries. Without using and writing SQL, business users can ask questions in NPL and get direct responses with Cortex Analyst.

Finally, let's discuss healthcare and pharma. This industry is highly regulated and rigid. However, there's a reason. Because we must ensure that drugs, whether patented or generic, are manufactured to the greatest standards. When implementing commercial data platforms and working with sales and marketing data, Snowflake is typically the first option. 70% of AI spending is directed toward marketing and sales, according to the "State of AI in Business 2025" publication. Nonetheless, the previously mentioned Cortex AISQL could help many life sciences organizations advance. This implies that there will be chances for supply chain management, clinical development, drug discovery, and real-world data in addition to commercial data to harvest the AI features provided by Snowflake.

I witnessed plenty of use cases; two of them piqued my imagination and admiration. The first is Pfizer's AI-enabled biomarker discovery.

Biomarker discovery is the process of identifying and validating molecular indicators, such as proteins, genes, or metabolites, that signal a biological process, disease state, or response to a treatment.

Can you picture the thousands of articles that researchers must sort through for each biomarker? From this slow, large-scale screening of samples using technologies like genomics and proteomics to identify potential candidates resulted in a non-biased and fast process, bringing Pfizer higher success trial rates. AI enablement in biomarker discovery improves precision medicine by uncovering biomarker signatures that are essential for early detection and treatment of diseases within vast and diverse datasets.

The last example blew my mind. Health Services and the University of Alberta, in Canada, created their own medical scribe tool on Snowflake. Saving tons of time for the ER doctors whose note-taking was automated, leaving them with more time to treat patients. The tool got a name, Jenkins, and it became like a medical assistant between the doctor and the patients. According to the doctors, the results are very promising; they can treat 3-4 more patients per shift with fewer administrative errors and less patient waiting time.

Jenkins in action

Main takeaways:

• Go visit one of the Snowflake summits if you get the chance. It's a great location to network, and you will always learn something.

• You get to know a lot of Snowflake's partners; it's like a shopping mall for solutions with demos.

• As soon as possible, you must line up for lunch. It may take some time for you to get your share.

• Implementing and maintaining a top-notch data platform requires a village, just like raising a child. For a seamless process and delivery, it's best to depend on partners like dbt Labs, Fivetran, and Matilion from the software side. The client testimonials indicate that a significant portion of them worked with consulting firms such as SDG Group, Aimpoint Digital, Deloitte, or Informatica. These are but a handful of the many examples. https://www.snowflake.com/en/why-snowflake/partners/all-partners/

• Cloud data platforms are undoubtedly evolving fast due to AI, and the most recent iteration appears to be a hybrid of Databricks and Snowflake. I don't mind if we go on that path! I also saw Sigma looking a bit Tableau-ish and Power BI-ish at the same time.

• Snowflake community is awesome, join the local user group and become a data hero! https://community.snowflake.com/s/

• AI will be a buzzword in the next years. However, Snowflake isn't just talking about it; actually achieving it. They are influencing the direction of data culture and platform technology with their Cortex AI features, apps, and marketplace.