Does your LLM-as-a-judge respond based on it's mood? If yes, this is what might be happening:
One of the biggest lessons I've learned while building LLM evaluation systems at Microsoft is this: A good evaluation metric isn't just accurate—it has to be reliable.
#LLMEvals#GenAI
Social media algorithms today are built to reinforce your thinking. If you think about growth, abundance, career, life, the algorithms will show you posts related to that. The art of manifestation is right in front of us.
We just have to choose the direction we want to take.
Anthropic just released a 4-hour course to getting a $500k AI engineering job:
00:15 - The right way to prompt Claude
33:21 - What makes Claude act dumber on your code
01:33:39 - How Anthropic use Claude every day
02:50:56 - The fix that makes Claude way smarter
This 4-hour Anthropic free course replaces about 10 paid engineering courses.
Watch it today, then read the step-by-step guide on building loops below.
Paying $20 for flight internet just to work with ChatGPT feels expensive. Also, I don't want to wait to get to a town from Yellowstone just to know what Bisons eat. Having a browser that brings the best of online and offline AI is a great solution.
For Week 1, I am solving the problem of "getting things done while on the go". I have always felt the joy of using AI should not be limited to having a fast and stable internet connection.
It's Monday and I am starting a new series on 52 Products in 52 weeks. I'll be building a new product each week solving a problem and will post about it here.
#buildinpublic#AI#ArtificialIntelligence