Do you know when each new model drops, it's benchmarks, and what it's particularly good at? Have experience applying different models to build workflows to get things done that a single model can't?
Want to work on a startup working with billions of tokens a day?
We're hiring an Applied AI Engineer at MathGPT, where you'll help over 4 million students go from struggling with math -> understanding it confidently.
We're Looking For
Extensive python experience (2-4+ years)
Understanding of different LLM strengths and LLM API features (ex. streaming and parsing structured outputs, predicted outputs, etc...)
Experience iterating on prompts, and other techniques to improve LLM performance (opportunities for SFT and RL model training)
Able to work with a research mindset - set up measurements, try hypotheses, and make progress on the end goal
Perks
Impact how millions of students learn math and how trillions of tokens get deployed in production
Competitive compensation ($)
Unlimited research tokens
