Rewriting All of Spotify's Code Base, All the Time

Abstract

We don't need LLMs to write new code. We need them to clean up the mess we already made.

In mature organizations, we have to maintain and migrate the existing codebase. Engineers are constantly balancing new feature development with endless software upkeep.

But what if you could rewrite your codebase, every single day, across thousands of repositories? What if your engineers didn't have to maintain their code?

At Spotify, we are seeing early success using LLMs to perform predictable, repeatable and effortless code migrations.

In this talk, we’ll share how we created an Agentic Migrator that has gotten over 3000 PRs merged across several engineering disciplines. We will tell you how we reason about solving the complexity of LLMs maintaining code at scale. From managing build feedback loops across thousands of repos, to evaluating prompt effectiveness and mastering the sheer complexity of our diverse codebase.


Speaker

Jo Kelly-Fenton

Engineer @Spotify

Hi! I am Jo – an engineer at Spotify, working on the development and company-wide adoption of autonomous coding agents. Before working at Spotify, I was working on integrating automation into Amazon’s Grocery Warehouse workflows.

Read more

Speaker

Aleksandar Mitic

Software Engineer @Spotify

I'm Aleksandar! I have been working as a Software Engineer at Spotify for 4 years. My tenure at Spotify has been spent within Platform teams. Specifically working on infrastructure for service to service communication and backend developer experience. Most recently I've been working on AI assisted development tooling.

Read more

Date

Tuesday Mar 17 / 11:45AM GMT ( 50 minutes )

Location

Churchill (Ground Fl.)

Slides

Slides are not available

Share

From the same track

Session AI/LLM

Reliable Retrieval for Production AI Systems

Tuesday Mar 17 / 10:35AM GMT

Search is central to many AI systems. Everyone is building RAG and agents right now, but few are building reliable retrieval systems.

Speaker image - Lan Chu

Lan Chu

AI Tech Lead and Senior Data Scientist

Session AI

Beyond Context Windows: Building Cognitive Memory for AI Agents

Tuesday Mar 17 / 02:45PM GMT

AI agents are rapidly changing how users interact with software, yet most agentic systems today operate with little to no intelligent memory, relying instead on brittle context-window heuristics or short-term state.

Speaker image - Karthik Ramgopal

Karthik Ramgopal

Distinguished Engineer & Tech Lead of the Product Engineering Team @LinkedIn, 15+ Years of Experience in Full-Stack Software Development

Session AI/ML

Refreshing Stale Code Intelligence

Tuesday Mar 17 / 01:35PM GMT

Coding models are helping software developers move faster than ever, but weirdly, the models themselves are not keeping up. They are trained on months-old snapshots of open source code. They have never seen your internal codebase, let alone the code you wrote yesterday.

Speaker image - Jeff Smith

Jeff Smith

CEO & Co-Founder @ Neoteny AI, AI Engineer, Researcher, Author, Ex-Meta/FAIR

Session applied ai

Building an AI Gateway Without Frameworks: One Platform, Many Agents

Tuesday Mar 17 / 03:55PM GMT

Early AI integrations often start small: wrap an inference API, add a prompt, ship a feature. At Zoox, that approach grew into Cortex, a production AI gateway supporting multiple model providers, multiple modalities, and agentic workflows with dozens of tools, serving over 100 internal clients.

Speaker image - Amit Navindgi

Amit Navindgi

Staff Software Engineer @Zoox

Session

Async Agents in Production: Failure Modes and Fixes

Tuesday Mar 17 / 05:05PM GMT

As models improve, we are starting to build long-running, asynchronous agents such as deep research agents and browser agents that can execute multi-step workflows autonomously. These systems unlock new use cases, but they fail in ways that short-lived agents do not.

Speaker image - Meryem Arik

Meryem Arik

Co-Founder and CEO @Doubleword (Previously TitanML), Recognized as a Technology Leader in Forbes 30 Under 30, Recovering Physicist