Summary

In this episode of the Overcommitted Podcast, hosts Bethany and Erika sit down with Warren Parad, CTO and co-founder of Authress, a user authorization API built for reliability. Warren shares how his team stayed fully operational during the massive AWS US-East-1 outage in October 2025 using DNS failover and multi-region strategies, and what the delayed alert logs taught them about timestamp trust.

The conversation kicks off with a candid discussion on AI agents and critical thinking, whether managing multiple coding agents is really multitasking or just micromanagement, and what the trade-offs mean for early-career engineers. Warren traces his reliability-first mindset back to his roots in electrical engineering and healthcare IT, where late-night on-call pages through Citrix proxies and hospital billing systems shaped how he thinks about uptime today. The group also explores what it really takes to build a Five Nines organization and how hiring practices need to match the reliability culture you want. The episode wraps up with a round of Never Have I Ever: SRE Edition, featuring Friday deploys gone wrong, blaming DNS, and discovering outages from customer tweets.

Links

Authress, Warren's company, user authorization API for software makers. The product he's building and wants to plug.

Adventures in DevOps Podcast, Warren's podcast, co-hosted with Will Button. 300+ episodes on DevOps, engineering leadership, and cloud architecture.

How When AWS Was Down, We Were Not, Authress's blog post detailing their resilience strategy during the October 2025 AWS outage. Referenced in Theme 1 questions.

So You Want to Build Your Own Authorization?, Warren's article on why authorization complexity creeps up on teams. Referenced in Theme 2 questions.

An Interview With Warren Parad, CIAM Weekly, March 2025 interview covering Warren's views on CIAM, FedCM, and the future of authentication.

FedCM, Browser Native Auth (Adventures in DevOps Episode), Adventures in DevOps episode diving into FedCM and why authentication should move from user-land to kernel-land.

Warren Parad on LinkedIn, Warren's LinkedIn profile.

Warren Parad on Bluesky, Warren's Bluesky profile.

Warren Parad on GitHub, Warren's GitHub profile, includes Authress repos, OpenAPI Explorer, and other open-source work.

Authress Knowledge Base, Technical articles from the Authress team on auth, security, and infrastructure.

Warren Parad, Personal Site, Warren's personal website.

Hosts

Overcommitted

Bethany Janos

Erika (Eggyhead)

Summary

In this episode of Overcommitted, Satish Saley, a senior software engineer with extensive experience on DoorDash's search platform team, discusses scaling search systems at a hyper-growth company. This conversation dives deep into software engineering challenges and software development strategies that impact programmer productivity and career growth. Satish details two major engineering transformations: rebuilding the search indexing pipeline with Kafka, Flink, and Elasticsearch, and later replacing Elasticsearch with a custom search engine based on Apache Lucene, which significantly improved performance and reduced costs.

The episode also explores the complexities of migrating off monoliths, securing leadership buy-in for technical rebuilds, and why respecting legacy systems is crucial in engineering culture. Additionally, the hosts share personal stories of database pivots and conclude with a confessional segment on technical decisions that didn't age well. This episode is packed with insights for software engineers and tech professionals interested in the intersection of technology, scaling, and work life balance.

Links

• Satish Saley on LinkedIn: https://www.linkedin.com/in/satish-saley-65527525/
• Conway's Law: https://en.wikipedia.org/wiki/Conway%27s_law
• Build Faster Indexing with Apache Kafka and Elasticsearch: https://careersatdoordash.com/blog/open-source-search-indexing/
• Introducing DoorDash's In-House Search Engine: https://doordash.engineering/2024/02/27/introducing-doordashs-in-house-search-engine/

• 
  

Hosts

• Overcommitted: https://overcommitted.dev
• Bethany Janos: https://github.com/bethanyj28
• Brittany Ellich: https://brittanyellich.com

Summary

In this special one-year anniversary episode of Overcommitted, Brittany, Erika, and Bethany reflect on their journey from a small accountability group to a thriving weekly podcast and community of over 160 software engineers. They discuss their experiences within the software development lifecycle, the challenges of maintaining programmer productivity, and the evolution of their engineering culture. Listeners will hear candid stories about mastering interview skills, handling the complexities of production admin, and navigating promotional efforts. The episode also highlights the podcast's unexpected role as a valuable networking tool for remote workers in tech careers. Looking ahead, the hosts share exciting plans including new panel-format episodes, community events, and automation initiatives. Closing with affectionate roasts in their first-ever Overcommitted Superlatives, this episode offers both insight and entertainment for anyone passionate about software projects and sustainable work-life balance in tech.

Links

• Episode 1 - Imposter Syndrome: https://overcommitted.dev/imposter-syndrome-in-software-engineering/
• Overcommitted Discord Server: https://discord.gg/d9gZyYuqKd
• Computer systems: A programmer's perspective: https://www.amazon.com/Computer-Systems-Programmers-Perspective-3rd/dp/013409266X

Hosts

• Overcommitted: https://overcommitted.dev
• Bethany Janos: https://github.com/bethanyj28
• Brittany Ellich: https://brittanyellich.com
• Erika Eggyhead: https://github.com/eggyhead