Cloud Architect

Werner Vogels Portfolio

Amazon CTO who has run the "All Things Distributed" blog since 2006, publishing detailed technical breakdowns of how AWS services like Lambda and S3 are architected internally, alongside broader distributed-systems design principles.

AWSDistributed SystemsSystem DesignTechnical Leadership

Location

Not listed — probably too busy shipping to update the bio.

Skills

AWSDistributed SystemsSystem DesignTechnical Leadership

What makes it work

A breakdown of the choices that make this portfolio stand out.

1

Nearly two decades of continuous publication under one identity

Running the same blog since 2006 under a real name and title gives the archive a rare quality: you can trace how his thinking on distributed systems evolved alongside the infrastructure he was building.

2

Internal engineering writeups made public

Posts on "the invisible engineering behind Lambda's network" translate what would normally stay an internal design doc into a public technical narrative, which is a different and harder skill than announcing a feature.

3

Architecture principles stated as durable rules, not trend commentary

The writing favors general distributed-systems principles over reacting to industry news, which is why older posts remain useful years after the specific service they describe has changed.

What Cloud Architects can take from this

Specific, actionable tips to apply to your own portfolio — no generic advice.

  • If you have access to how a system was actually built, write about the engineering decisions behind it, not just the feature announcement — that is the harder and more valuable piece to produce.

  • Sustain one blog under one identity for years rather than starting over on a new platform — continuity is itself part of the credibility.

  • State the general principle behind a specific design choice so the writing stays useful after the specific technology it describes has moved on.

Case studies

Amazon's CTO uses his blog as a primary-source archive: a 2007 systems paper, the internal launch doc for Lambda, and a cost framework with its own podcast. Each piece shows the reasoning behind AWS, not just the product.

Case StudyintermediatePublished November 14, 2024

AWS Lambda turns 10: the annotated PR/FAQ

AWS LambdaPR/FAQ working-backwards docsFirecrackerServerless

A 5,460-word post that publishes Amazon's original internal PR/FAQ for AWS Lambda, dated 'Nov XX, 2014', with Vogels' notes in the margins. The notes mark what changed between the doc and launch, such as billing going from 250ms to 100ms and now 1ms. They also explain the tradeoffs: Node-only at launch, and single-tenant EC2 instances until Firecracker.

Few engineers publish the original design doc next to what actually shipped. The notes trace a decade of architecture decisions back to the launch constraints.

  • The full 2014 internal PR/FAQ, placeholders like 'CTO of XXX' left in

  • Notes on billing granularity: 250ms planned, 100ms at launch, 1ms today

  • Explains why Lambda first ran on single-tenant EC2 before Firecracker microVMs

Projects

ResearchadvancedPublished October 2007 (SOSP '07)

Dynamo: Amazon's Highly Available Key-value Store

Consistent hashingVector clocksQuorum replicationAnti-entropy recoveryDistributed hash tables

The SOSP 2007 paper describing Dynamo, Amazon's internal key-value store built for incremental scalability and high availability. It combines DHTs, consistent hashing, versioning, vector clocks, quorums and anti-entropy recovery, and lets users trade off cost, consistency, durability and performance. Vogels hosts the full text on his blog.

It is a real production system written up for peer review, including the lessons learned. It shows he can explain distributed-systems tradeoffs at research depth.

  • Nine authors from Amazon.com; Vogels is listed last

  • In Vogels' words, 'the first production system to use the synthesis of all these techniques'

  • Full author's version plus PDF free on allthingsdistributed.com

ToolintermediateOngoing; podcast episode 7 published October 27, 2025

The Frugal Architect

Cloud cost architectureSustainabilityPodcastLong-form interviews

A site of seven 'Laws of Frugal Architecture' for cost-aware cloud systems, grouped into Design, Measure and Observe phases. Examples: 'Make Cost a Non-functional Requirement' and 'Unobserved Systems Lead to Unknown Costs'. It adds articles by practicing architects and a podcast hosted by Vogels.

Making cost a design requirement instead of a finance clean-up job is a skill that sets architects apart. The laws make it teachable and checkable.

  • Seven laws across three phases: Design, Measure, Observe

  • Podcast hosted by Vogels with named practitioners (Zillow, The Ocean Cleanup)

  • Linked from allthingsdistributed.com as 'Stories from Frugal Architects'

Ready to build your portfolio?

Follow the Cloud Architect roadmap — skills, projects, and timeline to get hired.

Cloud Architect Roadmap