Notes on building systems that don't fall over.
A working journal from the intersection of cloud architecture, software engineering, agentic AI, networking, and the craft of leading technical teams.
30 Years of Enterprise AI: Inside Thomson Reuters' AI Platform
Laura Skylaki, VP of AI Engineering at Thomson Reuters, on what enterprise AI maturity really looks like - 16 internal services, an evaluation-first culture, engineers who haven't hand-written code in months, and knowing when NOT to use an agent.
Latest
90 postsHow Prime Video's Engineers Manage Agents, Not Code
Lilia Abaibourova, Principal Product Manager leading AI-native transformation at Prime Video, on how her team went from early CodeWhisperer experiments to a world where engineers manage fleets of coding agents, PMs submit production code, and the two-pizza team is being reinvented.
The Delivery Gap: Why 96% of Your AI Code Is Waste
Brenn of Delivery Hero on why most companies fail to see returns from AI coding tools - generating code 10x faster means nothing if your verification infrastructure can't keep up. Measure cost per accepted change, not PRs created.
What are MCP apps and why should you care?
Luigi Pederzani, co-founder of Manufact (behind mcp-use), on MCP apps - the first official extension of the MCP protocol that lets servers return interactive UI inside AI chat interfaces like ChatGPT, Claude, and Amazon Quick.
CLOUD
Distributed systems, infrastructure, and the trade-offs of building at scale.
The Hard Lessons of Cloud Migration: inDrive's Path from Monolith to Microservices
inDrive's Sasha Lisachenko and Artem Gab share a four-year journey from a bare-metal monolith to a cloud-native, geo-distributed platform on AWS, including their migration from Redis to Valkey on ElastiCache.
Evolving Lambda: from ephemeral compute to durable execution
Michael Gasch on AWS Lambda Durable Functions - the checkpoint-replay model that brings resilient, long-running, stateful workflows natively to Lambda, and when to choose it over Step Functions.
12-Factor-app on containers - re:Invent2021
A practical look at the 12-factor app methodology with building microservices with containers
ENGINEERING
Craft, code quality, and the practices that keep systems maintainable.
How Prime Video's Engineers Manage Agents, Not Code
Lilia Abaibourova, Principal Product Manager leading AI-native transformation at Prime Video, on how her team went from early CodeWhisperer experiments to a world where engineers manage fleets of coding agents, PMs submit production code, and the two-pizza team is being reinvented.
The Delivery Gap: Why 96% of Your AI Code Is Waste
Brenn of Delivery Hero on why most companies fail to see returns from AI coding tools - generating code 10x faster means nothing if your verification infrastructure can't keep up. Measure cost per accepted change, not PRs created.
5 Quality Gates That Let You Ship 250% Faster with AI Coding Agents
Ryan Cormack of Motorway on the five quality gates - spec-driven planning, AI-assisted review, deterministic tests, complexity checks, and human review - that let 120+ engineers ship 250% more deployments without breaking production.
AGENTS
Autonomous agents, tool use, evaluation, and the emerging LLM stack.
30 Years of Enterprise AI: Inside Thomson Reuters' AI Platform
Laura Skylaki, VP of AI Engineering at Thomson Reuters, on what enterprise AI maturity really looks like - 16 internal services, an evaluation-first culture, engineers who haven't hand-written code in months, and knowing when NOT to use an agent.
What are MCP apps and why should you care?
Luigi Pederzani, co-founder of Manufact (behind mcp-use), on MCP apps - the first official extension of the MCP protocol that lets servers return interactive UI inside AI chat interfaces like ChatGPT, Claude, and Amazon Quick.
5 Lessons Running AI Agents in Production
John Sexton and Aaron Tummon of Genesys share five hard-won lessons from running an agentic AI copilot for 2 million users across 21+ AWS regions - framework choice, orchestration, context management, prompt caching, and relentless testing.
NETWORK
Protocols, latency, and the physical realities beneath the abstractions.
My Ultimate SD-WAN solution
What a cool SD-WAN solution could be
Riverbed SD-WAN Advanced Routing capabilities
Getting under the hood of Riverbed SD-WAN's routing capabilities
What if the future of networking was networkless? part 2
Thoughts about the future of Networking, the second part
LEADERSHIP
Growing teams, technical direction, and the human side of engineering.
Byte Club ep.4 ft Romain Jourdan from AWS
Romain Jourdan from Amazon Web Services (AWS) joins Byte Club episode 4 by MTM
2024 in review
Reflections on a turbulent year leading Developer Relations at AWS - navigating organizational change, an RTO mandate, and rebuilding trust - and the leadership lessons learned along the way.
Leading Distributed Teams in the Future of Work
Brian Elliott on leading distributed teams with intentionality - team-level agreements, context-sharing, and why regular in-person gatherings drive lasting engagement.