Feature engineering at scale

Feature platforms centralize how ML teams build, share, and serve features across training and inference. They promise to solve the chaos of scattered feature scripts, inconsistent data pipelines, and the dreaded “it works in training but fails in production” problem.

But building one that actually works at scale? That’s where it gets interesting. How do you handle real-time streaming, manage costs, and keep features fresh across multiple models and teams?

Join this session to share experiences and learn from others tackling the same challenges.

What we’ll explore together:

• Share your current setup: from cron jobs to streaming pipelines, every approach has trade-offs worth discussing
• Real-time vs near real-time: What does “real-time” actually mean for your use case? How fresh is fresh enough?
• Architecture experiences: versioning strategies, testing approaches, and monitoring solutions that have worked in practice
• Cost reality check: Compare notes on compute bills and storage strategies
• Scaling journeys: key learnings from moving to production scale and handling growth challenges

Bio

Siri is a backend engineer turned AI builder. She believes AI has the potential to reshape how we interact with technology - not by forcing people to adapt to machines, but by building products that naturally fit into everyday life.

At Nilenso, she has worked with a hyperlocal logistics company to build event-driven architectures, set up a data science platform, and improve developer testing tooling. Her experience spans building products and platform services that support millions of users per minute.

Outside of work, she’s a generalist who can often be found brewing coffee, logging 5Ks on nearby trails, or exploring new tech gear.