Skip to main content
Back to Home

Now

Last updated: June 2026 · Israel

A snapshot of what I'm working on, learning, and paying attention to right now. Inspired by Derek Sivers' /now movement.

Building

SRE Agent

A multi-agent AI system for autonomous cloud operations. Building the logic for an agent that can detect, diagnose, and respond to infrastructure incidents in real time — with minimal human intervention — across complex enterprise environments. Lots of interesting challenges around reliability, trust, and knowing when to escalate.

AI Native Organization Tooling

Building the guidance, tooling, and infrastructure to help engineering teams become genuinely AI-native — not just using Copilot for autocomplete, but rethinking how software gets built when AI agents are part of the team.

Learning

AI agent evaluation. How do you know your agent is actually good? Not just that it passed your test cases, but that it'll behave correctly at the edge cases production will inevitably surface. Working through evaluation frameworks, LLM-as-judge patterns, and what "correctness" even means for agentic systems.

Model Context Protocol (MCP). Anthropic's open standard for connecting AI models to tools and data sources is quietly becoming infrastructure. Exploring what a well-designed MCP server looks like and what the failure modes are.

Engineering leadership at scale. Reading deeply on how engineering organizations compound — or don't — as they grow. What separates engineering cultures that stay productive at 500 engineers from the ones that slow to a crawl.

Reading

  • AI EngineeringChip Huyen — The definitive guide to building production AI systems. Dense with practical insight.
  • The Pragmatic ProgrammerHunt & Thomas — Re-reading. It holds up.
  • Chip WarChris Miller — The geopolitics of semiconductors — essential context for anyone building on Sovereign AI.
  • Simon Willison's Weblogsimonwillison.net — The best single source for practical LLM engineering.
  • The Pragmatic Engineer NewsletterGergely Orosz — Required reading for engineering leaders.