Firehose

4 JUN 2026 · Podcast · Latent Space: The AI Engineer Podcast

Reality: The Final Eval — Lukas Petersson and Axel Backlund of Andon Labs

In this episode, Lukas Petersson and Axel Backlund from Andon Labs discuss their innovative AI evaluation benchmarks that focus on real-world agent performance, including their Project Vend vending machine business and multi-agent systems. …

AI evaluation benchmarks long-horizon agents multi-agent systems agent autonomy AI in business model aggressive behavior evaluation harness robotics benchmarks spatial reasoning agent observability OpenClaw AI alignment agent logging real-world AI deployment agent self-modification

Browse by tag

Reality: The Final Eval — Lukas Petersson and Axel Backlund of Andon Labs