Cheng Lian (@liancheng) 's Twitter Profile
Cheng Lian

@liancheng

Dad; Engineering at Databricks; Opinions are my own.

ID: 29218822

calendar_today06-04-2009 15:12:30

10,10K Tweet

2,2K Takipçi

442 Takip Edilen

Cheng Lian (@liancheng) 's Twitter Profile Photo

Vim served me well for 20+ years, but decided to switch to Neovim, world leader in plugin deliveries for better LSP support. After several nights fiddling with Lua configurations, settled down with LazyVim. So smooth. Having tears rolling in my eyes. Folke Lemaitre is my new hero. Hats off 🎩

Databricks (@databricks) 's Twitter Profile Photo

Credential vending for #UnityCatalog open APIs now in Public Preview. Benefits include: - Seamless interoperability across diverse compute engines - Simplified permissions management - A foundation for advanced governance features Learn more👇 dbricks.co/41NaFoe

Zhiding Yu (@zhidingyu) 's Twitter Profile Photo

Zizheng was one of our interns at NVIDIA back in summer 2023. Later, when we were considering to make him a FT offer, he chose to join DeepSeek without much hesitance. Back then, the DeepSeek multimodal team only has 3 people. I am still very much impressed by Zizheng’s decision

Cheng Lian (@liancheng) 's Twitter Profile Photo

Casually performed a 5B x 13M rows join using PySpark on Databricks serverless. The Databricks SQL query profile UI is now available for Python and Scala workloads. Goodbye, Spark UI!

Casually performed a 5B x 13M rows join using PySpark on Databricks serverless. The Databricks SQL query profile UI is now available for Python and Scala workloads. Goodbye, Spark UI!
Bohu (@bohutang) 's Twitter Profile Photo

Blog from 迟猫猫🐱, "Plan Representation: #1 Lesson Learned from Building an Optimizer." It's a fantastic read for anyone interested in database optimization. Excited for the next part!! skyzh.dev/blog/2025-02-0…

Cheng Lian (@liancheng) 's Twitter Profile Photo

Debugged a slow job containing 6 joins, each producing/shuffling 5.2B rows. It turned out that 4.3B rows were unnecessary due to a bad join condition. This streaming job runs every 5 min without crashing, for MONTHS. Databricks SQL is robust enough to be abused like that.

Debugged a slow job containing 6 joins, each producing/shuffling 5.2B rows. It turned out that 4.3B rows were unnecessary due to a bad join condition.

This streaming job runs every 5 min without crashing, for MONTHS.

Databricks SQL is robust enough to be abused like that.
Matei Zaharia (@matei_zaharia) 's Twitter Profile Photo

We’ve released a free edition of Databricks and opened up a lot of our training materials to help developers learn data engineering, data science and AI! Check it out to rapidly get started with the latest tools and with challenging use cases.

Cheng Lian (@liancheng) 's Twitter Profile Photo

I was using LSP4IJ to test my own language server, worked like a charm! I thought it leverages the LSP capability of IntelliJ Ultimate. Great to know that it also works for the Community edition.

Cheng Lian (@liancheng) 's Twitter Profile Photo

As long as major AI players still consider TUI as a major UI for vibe coding, Will McGugan's work on Rich and Textual is worth another $100M offer from Zucc. The attention and effort he pays to details, performance, and dev experience are phenomenal.

Cheng Lian (@liancheng) 's Twitter Profile Photo

Just embrace agile development and have a weekly release schedule. Now you restart your service once a week without feeling guilty.

Cheng Lian (@liancheng) 's Twitter Profile Photo

Installed Omarchy on my Framework 16 last night and luckily hit the uwsm outage on the first boot: github.com/basecamp/omarc… Love the result, though, totally worth the effort ❤️

Installed Omarchy on my Framework 16 last night and luckily hit the uwsm outage on the first boot: github.com/basecamp/omarc…

Love the result, though, totally worth the effort ❤️