Terence (T.J.) Brown

Data & Analytics Engineering. Opinions are my own.

[Yuchen Jin @Yuchenj_UW]

Super cool work from Databricks AI research team.

Data agents are harder than coding agents. Coding agents have verifiable tests. Data agents have to find “truth” across millions of tables, docs, dashboards.

Databricks Genie got to 91.6% accuracy, while the leading coding agent only got 32% on enterprise data analysis tasks.

Specialized knowledge search + Parallel Thinking + Multi-LLM is the key.

Databricks has an amazing research team, and I've been enjoying working with them!

[Terence (T.J.) Brown @SQLTJ]

Seems like a business decision took precedence over their users.

[Terence (T.J.) Brown @SQLTJ]

This is kind of embarrassing for the PBI team to explain to their users their decision making but still not address BI compatibility mode.

PBI users know that @tableau and @sigmacomputing want their products to work with UC Metric Views. Why cant PBI users also have nice things

[Christian Wade @_christianWade]

Important clarifying update to our product documentation posted regarding integration with 3rd party semantic models #PowerBI

https://t.co/E3h8wgRfBd

[Excess Returns @excessreturnpod]

"However much they're earning now, their revenues are only as durable as the spend from the person above them who is buying their products. And as you do get further down the layers, you do lose visibility in what's going on above you."

GMO's Tom Hancock on the layered nature of AI spending.

[Ben James @BenJames_____]

I made a USB-Clawd who gets my attention when Claude Code finishes a response

[Bazinga @ai_agi_asi_]

@Polymarket Microsoft products are so bad they successfully recreated this exact scene
WILD 😭

[Terence (T.J.) Brown @SQLTJ]

@ChrisDemahy @DanWahlin Tried it, it wasn’t enjoyable. CC and Gemini are both better.

[Terence (T.J.) Brown @SQLTJ]

@OmarShahine Sounds like a security match made I heaven. 🤣

[Terence (T.J.) Brown @SQLTJ]

@SQLGene Tbf, most users rarely use the interactivity to query the data, even after they’ve been showed how.

[Terence (T.J.) Brown @SQLTJ]

@kurtbuhler @ouraring @fitbit Are you pulling the raw data with the oura api?

[Terence (T.J.) Brown @SQLTJ]

@ltmundida @kurtbuhler All their competitors already have it.

[Terence (T.J.) Brown @SQLTJ]

@kurtbuhler @evanyi_81 Hmm, it seems like the governance and distribution for these types of data apps are largely solved in both Databricks and Snowflake.

[Lydia Hallie ✨ @lydiahallie]

Want to host Claude meetups in your city? We'll cover the funding, send swag, and give you monthly API credits for your demos.

You also get access to pre-release features and a private slack with the team! Go apply 💛

[Terence (T.J.) Brown @SQLTJ]

@bcherny How is this different from using the work tree plugin?

[Heinrich @arscontexta]

https://t.co/XmYSO3mC8f

[Prasenjit @Star_Knight12]

claude usage limit reached, your limit will reset at 7AM

[Terence (T.J.) Brown @SQLTJ]

@_catwu And…. It’s gone 🤣

[Machina @EXM7777]

turn this thread into instructions for your claude[.]md file

this might just change your life

[Tony Castricone @Castricone]

A 2-star QB scrambling for potentially the national-championship-winning touchdown on 4th down in his hometown against his hometown team after winning the Heisman for the most losing program of all time while his mother with MS screams for joy from her seat is possibly the limit of how good sports can get

[Terence (T.J.) Brown @SQLTJ]

@kurtbuhler Looks great 👍