yugoru

4 ms·

1. 1.▲
   by yugoru 5mo ago

   All my services have lost their databases. All their AI agent is offering me is to delete all the data (thousands of data) and create a new database. It insists there's no other way. But I have clients, I have hundreds of courses on th
2. 2.▲
   by yugoru 7mo ago

   its really interesting. Running agents close to the user’s data solves a lot of latency and privacy issues. The hard part seems to be balancing autonomy with predictability — once agents start chaining actions locally, debugging behavior be
3. 3.▲
   by yugoru 7mo ago

   its harder than it first appears. Even with good embeddings, semantic similarity across languages often breaks when articles include local context or idioms. Curious whether you found a threshold strategy that works reliably across language
4. 4.▲
   by yugoru 7mo ago

   Interesting perspective, one thing that keeps surprising me is how many modern systems still end up re-discovering OOP ideas in different forms, especially when you start modeling complex real-world systems (hardware pipelines, robotics con
5. 5.▲
   Do Startups Dream of Electric Robots(partenit.io)

   1 points by yugoru 1y ago|0 comments