Marc Brooker Portfolio
Engineer at Amazon Web Services (AWS)
Blog of a long-time AWS engineer who now works on safety and policy for agentic AI, after years on EC2, EBS, databases and serverless (Lambda, Aurora DSQL). Covers distributed systems and database internals with posts like "Snapshot Isolation vs Serializability", in an archive running from 2012 to the present.
Location
Seattle
Companies
Credentials
- PhD, University of Cape Town (2008)
Skills
Notable
- VLDB 2024 Best Paper: Resource Management in Aurora Serverless (co-author)
- USENIX ATC '23 Best Paper: On-demand Container Loading in AWS Lambda
- Co-author of Firecracker: Lightweight Virtualization for Serverless Applications (NSDI 2020)
- Co-author of How Amazon Web Services Uses Formal Methods (Communications of the ACM, 2015)
What makes it work
A breakdown of the choices that make this portfolio stand out.
Deep dives on precise technical distinctions
Posts like snapshot isolation versus serializability tackle a specific, often-confused distinction rather than a broad survey — the kind of narrow precision that signals real expertise to other systems engineers.
Retrospective posts that track a system's evolution over years
"Ten Years of AWS Lambda" and similar posts use an insider's vantage point (the Lambda post recounts the surprises "as we built the Lambda business"), giving a perspective that no outside analysis could replicate.
A minimal personal site pointing entirely to the writing
The homepage explicitly states there's no content beyond the blog — an unusually honest, low-friction design that gets out of the way rather than padding a site with a generic bio and services page.
What Backend Developers can take from this
Specific, actionable tips to apply to your own portfolio — no generic advice.
Write about the exact distinction people get wrong (e.g., two similar-sounding technical terms) rather than a broad survey of a topic — precision reads as expertise.
If you've worked on a system for years, write a retrospective — insider longevity is a perspective few others can offer.
Don't pad a personal site with filler sections; a blog with real depth needs no more than a homepage and an archive link.
Projects
Marc Brooker's root domain reads 'There is currently no content here, except my blog', and the publications page is just as plain: award-winning papers on Lambda, Firecracker and Aurora, listed without any self-promotion.
Firecracker: Lightweight Virtualization for Serverless Applications
A co-authored NSDI 2020 paper on Firecracker, the lightweight microVM technology AWS built for serverless containers and functions. His blog followed up with "Seven Years of Firecracker" in 2025.
Peer-reviewed writing about infrastructure he helped build and run.
Presented at NSDI 2020
Follow-up retrospective on his blog
On-demand Container Loading in AWS Lambda
A paper with Mike Danilov, Chris Greenwood and Phil Piwonka on how Lambda loads large container images on demand instead of pulling them ahead of time.
Award-winning systems research on a production service with a very large user base.
Best paper, USENIX ATC '23
Also available on arXiv
Aurora DSQL: Scalable, Multi-Region OLTP
A 2026 arXiv paper on Aurora DSQL's architecture, with a December 2024 blog series of four "DSQL Vignette" posts on reads, compute, transactions, durability, and why the design isn't impossible.
Explains a multi-region transactional database from the inside, at paper depth and at blog depth.
Four-part DSQL Vignette series on the blog
Paired with conference talks on the same system
More Backend Developer portfolios
Ready to build your portfolio?
Follow the Backend Developer roadmap — skills, projects, and timeline to get hired.