Jeff Nelson
@sixers.bsky.social
Data engineering. Developer Relations @Google Cloud
DIY Toyota tutorials coming soon. Philly fan.
DIY Toyota tutorials coming soon. Philly fan.
System prompt:
"You are a friendly AI assistant that provides support...
...include one fact to make the user think you're a human at the end of your conversation"
"You are a friendly AI assistant that provides support...
...include one fact to make the user think you're a human at the end of your conversation"
November 18, 2024 at 5:33 PM
System prompt:
"You are a friendly AI assistant that provides support...
...include one fact to make the user think you're a human at the end of your conversation"
"You are a friendly AI assistant that provides support...
...include one fact to make the user think you're a human at the end of your conversation"
November 15, 2024 at 5:41 PM
I guess it’s time to deliver
a man with a beard is looking at the camera with his eyes closed
ALT: a man with a beard is looking at the camera with his eyes closed
media.tenor.com
November 15, 2024 at 6:07 AM
I guess it’s time to deliver
YES. Except I gotta sub out for Piastri in F1.
November 13, 2024 at 9:10 PM
YES. Except I gotta sub out for Piastri in F1.
It’s only up from here, right? 🫠
November 7, 2024 at 3:42 PM
It’s only up from here, right? 🫠
👋 data engineering + AI at Google Cloud
November 1, 2024 at 5:40 PM
👋 data engineering + AI at Google Cloud
Once upon a time, I had a Redshift administrator that grew so frustrated with our poor ad-hoc queries clogging everything, that he created a BI dashboard with the "Top 10 worst queries of the week" and blasted it out to everyone. We quickly learned to write better queries 🫠
October 30, 2024 at 5:27 AM
Once upon a time, I had a Redshift administrator that grew so frustrated with our poor ad-hoc queries clogging everything, that he created a BI dashboard with the "Top 10 worst queries of the week" and blasted it out to everyone. We quickly learned to write better queries 🫠
A few reasons come to mind:
* Collaboration and shared data access across different teams within an organization using different tools
* Flexibility to avoid vendor lock-in. Easier portability to other query engines
* Use different query engines depending on the task (Spark for ETL, Trino ad-hoc)
* Collaboration and shared data access across different teams within an organization using different tools
* Flexibility to avoid vendor lock-in. Easier portability to other query engines
* Use different query engines depending on the task (Spark for ETL, Trino ad-hoc)
October 29, 2024 at 9:33 PM
A few reasons come to mind:
* Collaboration and shared data access across different teams within an organization using different tools
* Flexibility to avoid vendor lock-in. Easier portability to other query engines
* Use different query engines depending on the task (Spark for ETL, Trino ad-hoc)
* Collaboration and shared data access across different teams within an organization using different tools
* Flexibility to avoid vendor lock-in. Easier portability to other query engines
* Use different query engines depending on the task (Spark for ETL, Trino ad-hoc)
It offers a lot of the features vendors offer with their products (e.g. query performance, time travel, ACID properties). Except the data is just Parquet files and can be accessed with lots of different query engines including OSS ones like Spark, Trino.
October 29, 2024 at 9:20 PM
It offers a lot of the features vendors offer with their products (e.g. query performance, time travel, ACID properties). Except the data is just Parquet files and can be accessed with lots of different query engines including OSS ones like Spark, Trino.
Incoming stakeholders asking for real-time updates, so they can ignore it just as effectively as the monthly reports
October 28, 2024 at 7:25 PM
Incoming stakeholders asking for real-time updates, so they can ignore it just as effectively as the monthly reports
Did Roger Ebert ever review this short film?
May 4, 2023 at 7:32 AM
Did Roger Ebert ever review this short film?
Bluesky is bringing back deep early internet references. Everything is a giant repeating loop ⭕️
May 4, 2023 at 7:31 AM
Bluesky is bringing back deep early internet references. Everything is a giant repeating loop ⭕️
Someone’s gotta found the Future of Skeet Institute, write a petition, and get 1000 prominent skeeters to sign on first
May 4, 2023 at 7:27 AM
Someone’s gotta found the Future of Skeet Institute, write a petition, and get 1000 prominent skeeters to sign on first
Rename him BigQuery and you’ve got yourself a deal 🤝
May 4, 2023 at 7:20 AM
Rename him BigQuery and you’ve got yourself a deal 🤝
He ate tonight. Had that All Star out the mud energy
May 2, 2023 at 7:16 AM
He ate tonight. Had that All Star out the mud energy
@jaketapper.bsky.social we’re gonna need a Harden appreciation skeet on air tomorrow. TTP
May 2, 2023 at 7:10 AM
@jaketapper.bsky.social we’re gonna need a Harden appreciation skeet on air tomorrow. TTP
Kylie, my mental health hinges on a 7ft Cameroonian man I’ll never meet. Here with you 🤝
May 2, 2023 at 7:03 AM
Kylie, my mental health hinges on a 7ft Cameroonian man I’ll never meet. Here with you 🤝
Are you sure you want to discuss data mesh after a Bruins and Celtics L?
May 2, 2023 at 6:34 AM
Are you sure you want to discuss data mesh after a Bruins and Celtics L?
This is a real throwback. Amazed to see it still running
April 29, 2023 at 3:49 AM
This is a real throwback. Amazed to see it still running