www.seantallen.com

Spooky Action at a Distance

Mon, 18 May 2026 07:00:00 -0400

Non-locality is fine. Non-locality your tooling can’t surface is the problem.

Claude Jankowski

Sat, 16 May 2026 14:27:00 -0400

Claude ain’t Janks, but it will have to do.

Go Slow to Go Fast

Thu, 07 May 2026 02:00:00 -0400

Slowing down isn’t just allowed. It’s the faster path.

Water Is Wet

Wed, 06 May 2026 08:00:00 -0400

You get one life. Spend it accepting things as they are.

Don't Fuck It Up

Thu, 30 Apr 2026 08:00:00 -0400

Words of wisdom.

How Do I Know It's Working?

Mon, 27 Apr 2026 08:00:00 -0400

I test because I want to know what I’ve built is working. That’s very different than valuing testing.

Batteries Are Magic

Fri, 24 Apr 2026 08:00:00 -0400

I understand how batteries work. It doesn’t make them any less magical.

The OODA Loop

Wed, 15 Apr 2026 08:00:00 -0400

Your instructions to an LLM can quietly bake in the failure mode you’re trying to prevent.

Meet Murdock: The Wildcard in the Ensemble

Mon, 13 Apr 2026 08:00:00 -0400

Every good team needs its wildcard.

GC Made Fast

Sun, 12 Apr 2026 00:00:00 -0400

Most engineers think GC is slow. They’re wrong. Coordination is slow. Pony’s per-actor heaps and message-passing model eliminate the coordination that makes traditional garbage collectors a performance problem.

Agentic Personae: Code and Cognition from Spotlights to Identity

Fri, 10 Apr 2026 08:00:00 -0400

An attention focus tells an agent where to look. A persona tells it how to think.

Why Simple Workloads Find the Hardest Bugs

Wed, 08 Apr 2026 07:00:00 -0500

Marco Primi and I join David Wynn on Bug Bash to talk about workloads — what they are, why they matter for testing distributed systems, and how even a simple while true loop can expose bugs your unit tests miss.

An Ensemble of Claudes

Sat, 28 Mar 2026 08:00:00 -0400

Small differences in prompts produce wildly different outputs. You can fight that or you can use it.

Teaching Claude to Write Pony

Fri, 27 Mar 2026 12:00:00 -0400

I treated Claude like a junior engineer, taught it principles, and watched it grow. Two weeks later, it’s on my short list.

Parma for Mortals

Fri, 06 Mar 2026 14:00:00 -0500

The ideas behind Parma, explained like we’re sitting at a bar and you asked me what I’d been working on.

Peter Alvaro and Me

Tue, 03 Mar 2026 16:00:00 -0500

How Peter Alvaro’s lineage-driven fault injection clicked with me, and why I ended up at Antithesis.

Pony Programming Language with Sylvan Clebsch and Sean Allen -- Conversation #7

Sat, 21 Dec 2024 07:00:00 -0500

Sylvan Clebsch, Mike Shah, and I talk about how Pony came to be, walk through Pony Patterns, and discuss the language’s design.

Pony: High-Performance, Memory-Safe Actors with Sean Allen

Sun, 03 Nov 2024 07:00:00 -0500

Kris Jenkins and I dig into reference capabilities, Pony’s actor model, and how it implements garbage collection without stop-the-world pauses.

Confidential Container Groups: Implementing confidential computing on Azure container instances

Sat, 25 May 2024 00:00:00 -0400

Parma adds less than one percent overhead beyond the underlying TEE while ensuring a security invariant over all reachable states of the container group.

Parma: Confidential Containers via Attested Execution Policies

Sun, 10 Apr 2022 00:00:00 -0400

Lift-and-shift deployment of unmodified containers with strong security guarantees, using VM-level isolation and attested execution policies.

Unproven Tech Case Study with Sean Allen

Sat, 25 Jul 2020 07:00:00 -0500

How we picked Pony to build a sub-millisecond latency stream processor for Wall Street, and what happened next.

On Deny Capabilities for Safe, Fast Actors

Wed, 08 Jan 2020 00:00:00 -0400

A Papers We Love talk on how Pony uses deny capabilities to give you data-race freedom and fearless concurrency.

Adventures in Cgo Performance

Wed, 18 Sep 2019 00:00:00 -0400

Hard-won lessons about using Cgo in performance-sensitive code, from building the Go API for Wallaroo.

Data Corrupting Architectures We Know and Love

Wed, 18 Sep 2019 00:00:00 -0400

How our favorite single-machine data-access patterns are just as unsafe when we scale them out to distributed systems.

The Promises and Perils of Eschewing Distributed Coordination

Thu, 08 Aug 2019 07:00:00 -0500

A QCon NY 2019 panel covering the promises and perils of building distributed systems without coordination.

Things I'd like in my code management tool

Wed, 26 Dec 2018 13:31:27 -0500

My wish list for a code management tool that actually understands how I work.

Pat Helland and Me

Fri, 26 Oct 2018 00:00:00 -0400

Applying Pat Helland’s ideas from ‘Life Beyond Distributed Transactions’ to build stateful applications that scale almost infinitely.

Pony

Fri, 26 Oct 2018 00:00:00 -0400

Why we chose Pony to build Wallaroo, what made it work, and where its immaturity bit us.

Scale-Independent Python

Sat, 06 Oct 2018 00:00:00 -0400

Write your Python code without thinking about scale, then run it on a platform that handles scaling for you.

How Did I Get Here?

Tue, 22 Nov 2016 09:56:56 -0500

How we combined fault injection, message tracing, and auditing to build trust in a distributed stream processor.

Learning From Ada

Mon, 22 Feb 2016 10:42:22 -0500

How a younger me chose C over Ada, and what that choice taught me about how we pick technologies.

Of Debt and Decisions

Sun, 19 Apr 2015 11:09:15 -0500

Technical debt isn’t always the enemy. Sometimes it’s a deliberate trade-off that lets you move forward.

On Becoming a Better Teacher

Sun, 12 Apr 2015 11:24:38 -0500

How conversations with Kathy Sierra changed the way I think about teaching and writing Storm Applied.

Storm Applied: Behind the Scenes

Sat, 04 Apr 2015 11:28:12 -0500

Behind-the-scenes vignettes from writing Storm Applied — rejected subtitles, manuscript struggles, and The Jam.

An Acceptable Tool

Tue, 24 Feb 2015 11:31:27 -0500

Why I’d rather have a tool that’s good enough everywhere than one that’s perfect at one thing.