Speaker
Abstract
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. How we saw arising different "engineering flavors" and which are their responsibilities in building, scaling and maintaining data/ML products at global scale. Massimo will also deep dive into Bumble Inc MLOps stack, enriching it with lessons learnt and best practices.
Interview
Currently leading some of the smartest folks I have ever worked with in the Integrity & Safety and MLOps teams at Bumble Inc, the parent company of Bumble, Badoo and Fruitz. My teams' focus is both in designing, deploying and maintaining world-class ML solutions that keep our platforms safe while enabling and facilitating the work of the wider Data Science Team, implementing a centralized MLOps platform able to support the end-to-end lifecycle of ML projects at the company.
The level of maturity we reached at Bumble in our machine learning endeavors required years of work and some mistakes along the way. In this talk, I will go through some interesting concepts, definitions and approaches that worked well for us and that can be replicated with little effort by other practitioners facing similar challenges, especially while building a machine learning engineering function.
Senior engineers or engineering leaders interested in learning more about other teams' ML challenges, both the ones that already started a successful machine learning journey or the ones that are falling into common pitfalls along the way.
A good understanding of what it means to be a machine learning engineer in a successful ML team and which responsibilities they should have in order to have an impact. A list of common challenges and solutions together with tools and frameworks that worked well (and the ones that didn't!).
Topics
QCon New York 2023 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.
Part of the track
MLOps: Navigating the Terrain of Large-Scale Models Hosted by Bozhao (Bo) Yu Founder @BentoML.aiFrom the same track
Wednesday 14 June
10:35 Salon D Session ML Infrastructure Introducing the Hendrix ML Platform: An Evolution of Spotify’s ML Infrastructure Divita Vohra, Mike Seid The rapid advancement of artificial intelligence and machine learning technology has led to exponential growth in the open-source ML ecosystem. 11:50 Salon D Session Machine Learning Improve Feature Freshness in Large Scale ML Data Processing Zhongliang Liang Engineering Manager @Facebook AI Infra 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. 13:40 Carroll Gardens Unconference Unconference: MLOps 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. 14:55 Salon D Session AI/ML A Bicycle for the (AI) Mind: GPT-4 + Tools Sherwin Wu, Atty Eleti OpenAI recently introduced GPT-3.5 Turbo and GPT-4, the latest in its series of language models that also power ChatGPT. 16:10 Salon D Session MLOps Platform and Features MLEs, a Scalable and Product-Centric Approach for High Performing Data Products Massimo Belloni Data Science Manager @Bumble 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. 17:25 Salon D Panel Panel: Navigating the Future: LLM in Production Sherwin Wu, Hien Luu, Rishab Ramanathan 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.