Evals & Software Factories

Everyone’s been talking about software factories lately. But what does that actually mean in practice? And how do you know your factory is producing anything good?

Once agents take over the coding, the bottlenecks move somewhere else: environments, builds, CI, testing, and evals. As agent output scales, evals become the quality gate, the way you know whether what your agent's ship actually works. The teams pulling ahead aren’t only adopting agents. They’re building the infrastructure, and the evals, that make AI development work at scale.

Join the 3rd Autonomous Engineering meetup, where platform and AI leaders who run these systems today will share what works, what breaks, and where to begin.

Agenda

17:00 Doors open, food and drinks

17:30 Opening words

17:40 Evals for software factories

Adam Gold, Founder & CPO @ islo.dev

18:00 Training AI models to out-execute attackers

Lior Moshe, CTO @ Arithmetic

18:20 Short break

18:30 You Have Agents. Do You Have a Factory?

David Mark, VP Platform Engineering @ Incredibuild

18:50 Don’t build good evals, just build honest ones

Gad Benram, Co-Founder & CTO @ TensorOps

19:10 The path to 1 in Enterprise RAG Benchmark: Evaluating our retrieval harness

Regev Ben Simon, Co-Founder & CTO @ Flexor.ai

19:30 Good time!

Details

Location: AWS Experience, Menachem Begin Rd 121, Floor 28, Tel Aviv

November 4, 2026, doors open 17:00

Talks in Hebrew

Please bring an ID. It’s required at the entrance.

Join the Autonomous Engineering community on WhatsApp: https://chat.whatsapp.com/CZrhaCokDFI83ANmlyE7Yr