Optiver

Optiver has completed its acquisition of a majority stake in Northpool, marking an important step in our expansion into European energy markets.

Combining Northpool's expertise in power and gas trading with Optiver's technology and scale, we're excited about what's ahead.

Read more: bit.ly/4zoUPhF

At AI Engineer Shanghai, Xinghong Fang will share how we've built and scaled a multi-region gateway to keep #LLM workloads available when providers or regions fail.

He'll cover provider failover, protocol translation, observability and the #engineering tradeoffs involved.

If you're attending, join the talk and visit our booth to meet the Optiver team.

@aiDotEngineer | Talk details: ai.engineer/shanghai/2026/…

If we gave our AI agents the full data catalog, they would drown in it.

Agents need only the slice that matters for the task in front of them: the one table they can trust, what it means, and how they're allowed to use it.

On November 4 at CONTEXT 2026, our Global Head of Data Platform, Christophe Godefroy, will pass the mic to his AI agent and a DataHub agent to explain how we built a context layer for agents in partnership with @DataHubCloud.

Register for the free virtual event: lnkd.in/eTQWsyh4

@aiDotEngineer NYC brought together engineers working through some of the same questions we're tackling at Optiver: how to make AI useful, reliable and scalable across everyday engineering workflows.

One highlight was Noyan Tokgozoglu's session, "Scaling AI with P&L on the Line". He shared how we're building ways to evaluate new models and make them available quickly, while maintaining the flexibility to choose what works best for our engineering teams.

Here's a look back at our time in New York.

In trading, your model is part of the system it's trying to predict.

Denis Perevalov, Quantitative Researcher, shares what makes ML in trading an interesting research problem. Read more ⤵️ bit.ly/4yURFls