Skip to content

Details

If English is the new programming language and everyone becomes a developer, what tools, workflows, and best practices will shape the next generation of ML systems?

In this session, we’ll survey the current state of ML developer tooling through the lens of the last five years, explore emerging best practices, and discuss what’s practically next. From prompt management and continuous evaluation to agent debugging and safety guardrails, we’ll look at the gaps in today’s workflows and the tools new developers will need to work productively with LLMs and agents.

We’ll also unpack the limitations: reliability, cost, latency, context, and autonomy — and how future tooling might make agents trustworthy collaborators rather than brittle prototypes.

Who should join: ML engineers, product builders, infra folks, researchers, and anyone curious about the evolving developer experience when “English is code.”

Members are also interested in