Topic
Posts on Data Engineering
3 min read Suyash Shrestha
Putting an LLM into production is mostly the boring parts
The demo is the easy bit. Making an AI feature reliable, affordable and honest is where the real work sits. Here is what I keep running into.
Read article
2 min read Suyash Shrestha
Long AI jobs need a proper tool, not a scheduled task and a prayer
Long, multi-step AI jobs fail partway through all the time. Here is why we use a tool built for that instead of stitching queues together and hoping.
Read article
2 min read Suyash Shrestha
Five signs your nightly data job should run live instead
Nightly data jobs are simple and cheap, right up until they are not. Here are five signs it is time to move to live data, and how to do it without starting over.
Read article