If you use it with OpenAI responses api, there’s not even any need to store input items in your own DB
Atlas has problem where it just gives up and quits after a few minutes, but Claude doesn't seem to have a time limit and will work through a batch of CDP codes successfully.
Hi HN,We built an AI agent for data analysts that turns the soul crushing spreadsheet & BI tool grind into a fast, verifiable and joyful experience. Early users reported going from hours to minutes on common real-world data wrangling tasks.It's much smarter than an Excel copilot: immutable data steps, a DAG of SQL views, and DuckDB for instant crunching over millions of rows. Our early...
I recently have started relying Google CloudShell as my remote development workspace. One thing I noticed is that all of my shells seem to come up in Google's Taiwan datacenter, so the latency to the US is horrible (100ms+ ping).Are you also experiencing this? You can check by `curl http://checkip.amazonaws.com` and then looking up the IP in a geoip database.
Native desktop client could utilize the same stack via Electron.
Backend could start with Python using a popular all-in-one framework like Django to begin with, and scale towards micro-service architecture over time as components grow in complexity.
Heroku is a good all-in-one infrastructure service that takes care of dev-ops for the first year.
If there really was patent infringement, then it would be difficult to erase all traces. Similar to the Zenimax Oculus lawsuit where computers were erased, but the verdict still came down against Oculus.
Are they going to fragment the code-base with Java to introduce a desktop client? Are they going to hire another enterprise Java dev, who most likely isn't familiar with the rest of the stack?
The rise of Electron is fundamentally driven by the cost effectiveness of the wide availability of developers and community support.
If you had a cloud based Varnish service that had a good web interface for defining simple/common caching rules, and also exposed advanced scripting, and had live debugging + logging, would you use it?Would you pay ~$40/month for an instance that had 10GB of SSD backed cache, 512MB of memory cache and 1TB of bandwidth? Would you pay more if it were actually two smaller instances that were...
From personal experience, my hypothesis is that in any organization, there are a handful of high impact but low cost improvements to make. These low-hanging fruits will not be blatantly obvious, and will not be discernible to someone who isn't "awesome" for whatever reason, e.g. lack of curiosity, incompetence, bad business sense, etc... In a startup with sufficient autonomy, it's probably even doable to make a lasting difference in the first week.
This is probably also tied to the idea of a 10x engineer, because they can figure out the right problems to solve and use 100% of their time to provide 1000% value.