MLOps: Navigating the Terrain of Large-Scale Models

The MLOps landscape is increasingly focused on deploying and managing large-scale models. This shift has revealed new prospects for innovation while concurrently posing distinct challenges. In this track, we delve into these nuances, shedding light on leveraging MLOps principles to maximize the potential of large-scale models and address the inherent complexities.

We will explore an array of strategies, from developing AI-driven applications and optimizing large-scale data processing to revolutionizing machine learning infrastructure. We bring these themes to life by showcasing tangible case studies from industry pioneers, including OpenAI, Meta, Spotify, and Bumble.

Key discussion points will include:

  • The crafting of high-functioning AI applications utilizing OpenAI's API and Plugins
  • Meta's strategies to invigorate feature freshness in the large-scale ML data processing
  • A walkthrough of Spotify's Hendrix ML platform as a progressive step in their ML infrastructure evolution
  • Bumble's scalable, product-focused approach to designing platforms and features for high-performing data products
  • Strategies for fostering collaboration and communication within MLOps teams tasked with large models

Irrespective of your role as a data scientist, machine learning engineer, or software developer, this track offers actionable insights for integrating large-scale models in your operations. Immerse yourself in our exploration of scalable machine-learning operations and draw from proven strategies to succeed in this rapidly evolving space.

From this track

Session ML Infrastructure

Introducing the Hendrix ML Platform: An Evolution of Spotify’s ML Infrastructure

Wednesday Jun 14 / 10:35AM EDT

The rapid advancement of artificial intelligence and machine learning technology has led to exponential growth in the open-source ML ecosystem.

Speaker image - Divita Vohra

Divita Vohra

Senior Product Manager @Spotify

Speaker image - Mike Seid

Mike Seid

Tech Lead for the ML Platform @Spotify

Session Machine Learning

Improve Feature Freshness in Large Scale ML Data Processing

Wednesday Jun 14 / 11:50AM EDT

In many ML use cases, model performance is highly dependent on the quality of the features they are trained and inference on. One of the important dimensions of feature quality is the freshness of the data.

Speaker image - Zhongliang Liang

Zhongliang Liang

Engineering Manager @Facebook AI Infra


Unconference: MLOps

Wednesday Jun 14 / 01:40PM EDT

What is an unconference? An unconference is a participant-driven meeting. Attendees come together, bringing their challenges and relying on the experience and know-how of their peers for solutions.

Session AI/ML

A Bicycle for the (AI) Mind: GPT-4 + Tools

Wednesday Jun 14 / 02:55PM EDT

OpenAI recently introduced GPT-3.5 Turbo and GPT-4, the latest in its series of language models that also power ChatGPT.

Speaker image - Sherwin Wu

Sherwin Wu

Technical Staff @OpenAI

Speaker image - Atty Eleti

Atty Eleti

Software Engineer @OpenAI

Session MLOps

Platform and Features MLEs, a Scalable and Product-Centric Approach for High Performing Data Products

Wednesday Jun 14 / 04:10PM EDT

In this talk, we would go through the lessons learnt in the last couple of years around organising a Data Science Team and the Machine Learning Engineering efforts at Bumble Inc.

Speaker image - Massimo Belloni

Massimo Belloni

Data Science Manager @Bumble


Panel: Navigating the Future: LLM in Production

Wednesday Jun 14 / 05:25PM EDT

Our panel is a conversation that aim to explore the practical and operational challenges of implementing LLMs in production. Each of our panelists will share their experiences and insights within their respective organizations.

Speaker image - Sherwin Wu

Sherwin Wu

Technical Staff @OpenAI

Speaker image - Hien Luu

Hien Luu

Sr. Engineering Manager @DoorDash

Speaker image - Rishab Ramanathan

Rishab Ramanathan

Co-founder & CTO @Openlayer


Wednesday Jun 14 / 10:30AM EDT


Track Host

Bozhao (Bo) Yu


Bo is a seasoned founder with a wealth of experience in mobile consumer, gaming, and ML infrastructure. He founded BentoML, an ML platform that powers thousands of organizations across the globe in both the public and private sectors. Bo has also created the most engaged MLOps developers community, increasing active members by 4X within a year, as well as the world's first iPhone sports fan in-stadium app with more than 70% engagement rate.

Read more
Find Bozhao (Bo) Yu at: