tldr;
I took 100 job postings and ranked them against my resume, my yearly reviews, and my LinkedIn recommendations using Jev to see how it performed.
What is Jev?
Jev is a different type of AI model. According to their website:
Jev is TypeSafe’s flagship model and the first System One model. System One models are built to make fast, structured decisions that software can use directly. Jev evaluates typed questions against a state and returns structured results directly. No text generation, no parsing. You get typed values and probability distributions that your code can branch on, sort by, and route with. Choice and Score also return confidence, which your code can use to decide whether and how to act on an answer.
There are three types of questions that Jev can help answer:
- Choice: Pick one option
- Score: Place on a scale
- Noul: Is this true?
Ranking 100 Job Postings
Based on the model’s capabilities I had Claude throw it a little test using their Python SDK. Essentially what I wanted was simple.
For every job posting, give me a boolean “good fit” boolean (Noul), a match score (Score), processing time, and tokens used. The report aggregated this and also told me the approximate cost to run such a thing. The result is pretty impressive:
I created an anonymized report if anyone is curious about the entire results.