Stephen Batifol, data scientist at Wolt, shares insights on building an ML platform, developer relations, and creating a thriving internal community. They discuss the challenges of onboarding data scientists, importance of documentation, simplifying the developer experience, and expanding services. They also touch upon MLflow, Qflow, observability, training models with multiple countries, building trust through feedback, and attracting talent through talks and content sharing.
Read more
AI Summary
AI Chapters
Episode notes
auto_awesome
Podcast summary created with Snipd AI
Quick takeaways
Building an internal ML platform at Vault prioritizes the needs of data scientists and fosters a community of knowledge sharing.
Meetups play a significant role in fostering the tech community in Berlin and attracting tech talent to the city.
Deep dives
Creating an ML Platform at Vault
One of the key insights from the podcast episode is the discussion around creating an internal ML platform at Vault. The guest, a former ML platform engineer turned internal developer advocate, talks about the importance of advocating and supporting the use of the platform by teams and data scientists. The platform is designed to make the data scientists' job easier by providing a range of features, such as dynamic workflows, resource allocation, and easy deployment options. The guest also emphasizes the significance of community building within the platform, fostering knowledge sharing and supporting data scientists in their work.
Berlin's Tech Scene and Meetups
The podcast episode also touches upon the tech scene in Berlin. The speaker highlights Berlin's status as a growing tech hub and discusses the importance of meetups in fostering a vibrant tech community. The guest shares their experience of organizing the ML Screen meetups, which have contributed to the revitalization of the tech meetup culture in Berlin. They touch upon the appeal of Berlin as a tech hub and mention the influx of tech talent to the city. Overall, the podcast offers insights into the thriving tech scene in Berlin and the role of meetups in driving community engagement.
Challenges of Creating a Machine Learning Platform on Kubernetes
Another key point discussed in the podcast episode is the challenge of building a machine learning platform on Kubernetes. The speaker highlights the difficulty of abstracting away the complexities of Kubernetes for data scientists who may not be familiar with the infrastructure. They delve into issues such as resource allocation, data management, and the learning curve involved in understanding Kubernetes concepts. The platform team acknowledges the importance of ensuring a smooth developer experience and addresses these challenges by providing tools and simplifying the platform's interaction with Kubernetes. The discussion sheds light on the hurdles faced when leveraging Kubernetes for machine learning workloads.
Bridging the Gap Between SREs and ML Platform
The podcast episode also touches upon the collaboration between ML platform teams and SREs. The guest highlights their involvement in the SRE squad and the close relationship between the two teams. They discuss the importance of ensuring the reliability and availability of the ML platform, highlighting the use of SLAs and on-call rotations. The discussion emphasizes the need for a joint effort between ML platform teams and SREs to maintain the platform's stability and performance. This collaboration showcases the integration of SRE practices within ML platform development, ensuring the smooth operation of critical machine learning workflows.
MLOps Coffee Sessions #178 with Stephen Batifol, Building an ML Platform: Insights, Community, and Advocacy.
// Abstract
Discover how Wolt onboard data scientists onto the platform and build a thriving internal community of users. Stephen's firsthand experiences shed light on the importance of developer relations and how they contribute to making data scientists' lives easier. From top-notch documentation to getting-started guides and tutorials, the internal platform at Wolt prioritizes the needs of its users.
// Bio
From Android developer to Data Scientist to Machine Learning Engineer, Stephen has a wealth of software engineering experience at Wolt. He believes that machine learning has a lot to learn from software engineering best practices and spends his time making ML deployments simple for other engineers. Stephen is also a founding member and organizer of the MLOps.community Meetups in Berlin.
// MLOps Jobs board
https://mlops.pallet.xyz/jobs
// MLOps Swag/Merch
https://mlops-community.myshopify.com/
// Related Links
--------------- ✌️Connect With Us ✌️ -------------
Join our slack community: https://go.mlops.community/slack
Follow us on Twitter: @mlopscommunity
Sign up for the next meetup: https://go.mlops.community/register
Catch all episodes, blogs, newsletters, and more: https://mlops.community/
Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/
Connect with Stephen on LinkedIn: https://www.linkedin.com/in/stephen-batifol/
Timestamps:
[00:00] Stephen's preferred coffee
[00:32] Takeaways
[01:35] Please like, share, and subscribe to our MLOps channels!
[03:00] Creating his own team!
[04:44] DevRel
[06:32] The door dash of Europe
[11:28] Data platform underneath
[12:55] Cellular core deployment uses open source
[14:21] Alibi
[16:08] Kafka
[16:59] Selling points to data scientists
[20:05] Language models concerns of data scientists
[22:12] Incorporating LLMs into the business
[23:55] Feedback from data scientists and end users
[27:37] User surveys
[30:11] Evangelizing and giving talks
[35:25] Tech Hub Culture in Berlin
[38:38] Kubernetes lifestyle
[42:55] Interacting with SREs
[45:28] Wrap up
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode