Suggestions
Rishabh Bhargava
Software and ML - MLOps Roundup
Rishabh Bhargava is the Co-Founder and CEO of Refuel.AI, a company focused on providing clean, labeled data at the speed of thought.23 He has been serving as the Editor at MLOps Roundup since 2020, exploring the challenges of implementing machine learning in the real world.13
Bhargava has a diverse background in technology and data science:
- He previously worked as a Staff Software Engineer at Cloudera in 2021.13
- From 2019 to 2021, he was in Sales Engineering at Datacoral Inc.3
- He served as a Machine Learning Engineer at Primer AI from 2017 to 2019.13
- He has academic experience as a Graduate Teaching Assistant at Stanford University and as a Flight Test Director and Research Assistant at the Stanford Intelligent Systems Lab.3
Bhargava's educational background includes:
- A Master's Degree in Computer Science from Stanford University.3
- A Year Abroad in Computer Software Engineering from the Massachusetts Institute of Technology.3
- A Master of Engineering (MEng) degree from the University of Cambridge.3
His LinkedIn profile (username: bhargavarishabh) showcases his extensive experience in the fields of machine learning, data science, and artificial intelligence.2
Highlights
For all the progress in data science, one of the most stubborn problems that’s persisted has been resume parsing.
Resume parsing can be complex — major variations in what titles/skills mean, discrepancies in file format, and jargon changing between industries.
In fact, a recent study found that traditional ATS algorithms and rules-based parsers were only able to attain 60-70% accuracy, leading to talent mismatch, lost opportunities, and wasted effort.
We put Refuel and an LLM based approach to the test, and realized higher accuracy (95% vs 60-70%), significant time/cost savings, and a flexible output schema.
Here’s how we did it ⬇
This is huge.
For the first time, we have an open-source model that’s state-of-the-art and outperforming GPT-4o, on most evals.
Today, Meta announced the release of the Llama 3.1 set of models, including Llama 3.1 405B, their largest open-source model to date.
This is a paradigm shift for a few reasons:
- Sam Altman was indeed correct when he said that the cost of intelligence is going to 0.
There are multiple competing providers for that intelligence, democratized for everyone.
-
You can now customize and manage the models/weights/infra yourself without compromising on performance, with a super-permissive license.
-
TCO is typically lower — assuming you manage this well, you can get cheap hardware and deploy on your own, so no premiums to be paid to any model providers
-
You get the benefit of complete ownership and security over your data.
We’ve already begun spinning up Llama 3.1 into the Refuel platform, and plan on making it available to all of our customers soon.
What excites you the most about the Llama 3.1 release?
