Pierson Marks is a software engineer, designer and Jellypod cofounder and CEO. He created SpeechSDK, a TypeScript library that gives applications a common interface to multiple text-to-speech providers, and writes about operating agent-assisted development workflows.[2][1]
Start with Building a software factory entirely on a Claude subscription →
Links
Website
X · piersonmarks 
GitHub · piersonmarks Cofounder and CEO · Jellypod
Career
Jellypod began with a personalized podcast app released in December 2023. The company launched its creator studio in January 2025 and later added video, voiceover and short-form products. Its production workflow generates scripts, audio and visuals while leaving creators controls for editing, casting voices and publishing.[1]
Contributions
Separate work discovery from implementation
Marks’s software-factory account uses Linear as the shared record of work. Scheduled routines turn health checks and feedback into issues; a separate label-triggered path sends a specific issue to an implementation routine. A small webhook service supplies the authenticated request. The distinction lets people control which proposed changes become implementation tasks and keeps browser verification and pull-request creation inside a repeatable workflow.[3]
Switch speech providers through one interface
SpeechSDK accepts text, a model identifier and a voice, then translates the request into each provider’s API. Provider implementations handle differences while applications receive a common audio result. It also normalizes delivery tags where supported. The abstraction reduces the application code changed when evaluating or switching speech models; provider-specific options remain available.[4]
Tools, writing & talks
Writing, research & talks
Start here
Building a software factory entirely on a Claude subscriptionpiersonmarks.comA concrete issue-tracker and webhook workflow for scheduled triage and agent implementation.
Start with the boundary between proposed work and implementation. You need familiarity with issues, webhooks and coding agents to reproduce it; adopt one small loop before connecting more routines.
Author[3]A personal software factory system built to automate software creation workflows with simplicity as a core principle. The creator shares how it works, what they learned, and how others can replicate it in an afternoon. It's a practical guide-plus-system for anyone looking to systematize AI-assisted software development.
Catalog workOriginal source- Building SpeechSDKpiersonmarks.com
Explains the common provider interface and audio output design.
Author and library creator[4]