Suggestions
Christian Griset
ML engineering at Stripe
Christian Griset is a highly accomplished professional with a strong background in physics and extensive experience in engineering, data science, and quantitative trading. He holds a Bachelor of Science (B.S.) with honors in Physics from Caltech and a Master's degree in Physics from UC Santa Barbara. Christian has held key positions at various renowned organizations, including being the Founder at Stealth, ML Engineering Lead at Stripe, and contributing to data science and data infrastructure projects at Clever Inc. He has also worked as a Quantitative Trader at KCG Holdings, Inc. and as an Assistant Trader at Susquehanna International Group, LLP (SIG). Moreover, Christian has a research background, having been an NSF Graduate Research Fellow at the University of California at Santa Barbara.
Highlights
Jim Simons was a personal hero of mine, widely regarded as history's most brilliant self-made billionaire. It hasn't been since Benjamin Franklin that anyone contributed so profoundly to math, science, the broader economy and philanthropy.
I hold so dear what he said about the success of the storybook Medallion fund, the world's greatest money-making financial engine: there's no secret sauce. He described its success as coming from a combustion engine of accumulated insights from top-tier talent.
It's easy to say, but impossibly hard to replicate. The greatest long-term impact doesn't result from one brilliant person's singular insight. Rather, it comes from the compounded work of a group of brilliant, tirelessly working minds striving for a common goal.
RAG is promising, but today they just don't work as well as promised.
RAG can be the essential link for building AIs tailored to your business, yet there’s broad disillusionment as teams struggle making RAGs production-quality. Even more, they’re typically at a loss for why their RAGs don't work and the next steps to take.
Over the past few months, we've spoken with numerous teams and AI experts about their experiences with RAG. We've distilled these conversations into a post demystifying how Vanilla RAG fails due to compounding effects of semantic dissonance. We also include actionable tactics and high-ROI strategies for bridging the gap to production readiness.
Have you encountered similar challenges with RAG systems? What strategies have worked for you? Share your thoughts, and feel free to reach out. Let's drive this conversation forward.
https://lnkd.in/gfhq6F3n
