DEV Community

kestrelfield profile picture

kestrelfield

I test AI models before I trust them. Plain-words posts on calibration, checking AI tools on real work, and what new AI releases mean in practice.

Joined Joined on 
An 8 million parameter model that asks two questions about anomalies

An 8 million parameter model that asks two questions about anomalies

Comments
6 min read

Want to connect with kestrelfield?

Create an account to connect with kestrelfield. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
How FastAPI turns type hints into a checked, documented API

How FastAPI turns type hints into a checked, documented API

Comments
6 min read
How Ollama runs AI models on your own computer

How Ollama runs AI models on your own computer

Comments
6 min read
Rogue AI agents, or a sandbox left open? What security experts say

Rogue AI agents, or a sandbox left open? What security experts say

Comments
7 min read
How LiteLLM works: one AI gateway in front of 100+ providers

How LiteLLM works: one AI gateway in front of 100+ providers

Comments 1
6 min read
Decision models, in plain words: AI that picks instead of writes

Decision models, in plain words: AI that picks instead of writes

Comments
3 min read
How to Check if an AI Model's Confidence Can Be Trusted

How to Check if an AI Model's Confidence Can Be Trusted

Comments
3 min read
Testing AI Agents on Real Work: What the Ironclad Note Points To

Testing AI Agents on Real Work: What the Ironclad Note Points To

Comments
2 min read
loading...