AI in Indian retail banking: fraud detection at the core
How India's banks are moving from rule-based alerts to shared, real-time AI fraud prevention, and where the gaps remain.
- AI
- banking
- fraud
- India
Blog
Notes from the sky
Notes from running SkyRealm: how things are built, what broke, and how we fixed it.
KV cache and prompt caching: a developer's guide
What the KV cache is, how prompt caching builds on it, and how to use both in your own applications.
- AI
- LLMs
- KV cache
umbrelOS on systemd-nspawn: the setup story
How we got umbrelOS running inside a systemd-nspawn container, step by step.
- umbrelOS
- containers
Who owns your AI model? Proving model lineage in the age of LLM theft
Three techniques that let you prove an AI model is yours, even after someone fine-tunes it and renames it.
- AI
- security
- LLMs