Tag

#productionml

3 posts0 followers

Mohit VermaMar 29 · 6 min read

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do No 500 errors. No latency spikes. Just 91% of production LLMs quietly degrading — and your dashboards showing green the whole time. Here's the core tension I ke...

aiwithmohit.hashnode.dev

Mohit VermaMar 29 · 2 min read

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do Your LLM is returning HTTP 200. Dashboards are green. And your model has been quietly degrading for 3 weeks. No error codes. No latency spikes. Just wrong answe...

aiwithmohit.hashnode.dev

Mikiko BazeleyDec 19, 2022 · 10 min read

A Discussion: Top Challenges of Moving Data Science to Production

Introduction We know MLOps and Production ML are hard. But what are the common challenges that plague teams and practitioners (regardless of industry or background) when it comes to deploying data science models to production? Sabrina Aquino from Sha...

mikiko.hashnode.dev

Search Hashnode

Trending tags this week

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do

Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do

A Discussion: Top Challenges of Moving Data Science to Production