Case studies from the GenAI frontier, scaling ML teams, from biologist to machine learning consultant- Erik Gafni - The Data Scientist Show #082 episode artwork

EPISODE · Feb 24, 2024 · 1H 3M

Case studies from the GenAI frontier, scaling ML teams, from biologist to machine learning consultant- Erik Gafni - The Data Scientist Show #082

from Daliana's Game · host Daliana Liu

Erik Gafni builds AI systems and teams. He founded Eventum AI (https://bit.ly/eventum-ai), an ML consulting company working with high-growth startups. We talked about GenAI projects he worked on, how he built production ML systems, how to scale ML teams, and his journey from biologist to ML researcher. Interested in working with Erik: https://bit.ly/erik-consulting Erik's LinkedIn: https://bit.ly/erik-gafni-LI(00:00:00) Introduction(00:01:59) Is GenAI overhyped?(00:04:28) Ascent translation with AI(00:11:58) Social media app with AI(00:14:00) Stable diffusion model evaluation(00:15:57) "Consult-to-hire" model(00:17:35) AI in biotech(00:22:46) Self-supervised learning(00:31:22) How he hires people(00:33:19) Research vs production(00:35:57) Is AGI coming?(00:37:30) New trends in GenAI(00:41:45) Data quality in GenAI(00:42:58) Philosophy in LLMs(00:49:48) OpenAI vs Open Source(00:53:58) Mistakes he made(00:57:41) How did he get into ML

Episode metadata supplied by the publisher feed · Published Feb 24, 2024

Embed this episode

NOW PLAYING

Case studies from the GenAI frontier, scaling ML teams, from biologist to machine learning consultant- Erik Gafni - The Data Scientist Show #082

0:00 1:03:40

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of Daliana's Game?

This episode is 1 hour and 3 minutes long.

When was this Daliana's Game episode published?

This episode was published on February 24, 2024.

Can I download this Daliana's Game episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!