Plus editable axes, reference lines, and a bunch of quality of life improvements that make building faster and smoother. Really excited to see what our users build with these new tools.

We just rolled out an upgrade to our Charts experience! You now get access to a whole new library of visualizations - sankey, chord, waterfall, tornado, stacked charts, and more.

https://t.co/hQpICzqfnW works like having a data expert in your pocket Teams scale more easily, access data faster, and spend less doing it. Less friction between teams leads to better decisions across the business

Now imagine a single place where analytics teams build and govern data, and business users consume it. Thre data is consistent, the numbers are trusted, and access is fast

Business users need faster access to data, confidence in the numbers, and answers they can act on immediately.

Analytics teams need the freedom to build pipelines quickly, adapt governance, and keep up with endless ad-hoc requests.

Analytics teams and business users are working toward the same goals, just from different platforms.

and yes, SOC 2 Type 2 is in progress. privacy isn’t a feature, it’s the architecture.

with respect to the agent security, every action is transparent, explainable, and logged. you approve before anything runs. workspaces stay isolated, and all LLM activity is monitored and auditable.

now LLM training & data access, your data never goes to the LLM. the model never sees your values, only sanitized schemas. no external tools, only our own secure in-house ones.

let's start with data retention, we don’t store your data unless you opt in. logging out wipes everything. only SQL code and message history are retained. delete something and the agent won’t remember it.

lots of questions about data security lately, so here’s how Yorph AI is built to protect your data

We’ve built in best practices and abstracted the complexity so you can stay focused on building, just by asking.

All this is wrapped into an experience that feels simple. Tell Yorph the workflow you want, it happens automatically.

GCP is the backbone. Cloud Run, Pub/Sub, BigQuery, GCS, a reliable, secure, fully cloud-native setup. Nothing weird, nothing experimental.

DBT keeps everything clean. Our AI works iteratively with dbt to ensure SQL is correct, version-controlled, and always compiles.

ADK runs our agents. We use Google’s agent framework to orchestrate a multi-agent system that handles different tasks behind the scenes.

We’ve been asked a lot lately, “What’s actually powering your AI?” Quick rundown, no architecture diagrams needed.

Always great to see https://t.co/hQpICzqfnW mentioned in the wild. https://t.co/09trrh8TkS Thanks for the shout-out in this article!!

We just shipped database connectors!!! Definitely one of our biggest challenges we've worked on. Safe dry runs, full dataset warehouse executions etc. Yorph's first database connector is now live at https://t.co/hQpICzqfnW !!