👋 Hey there, I’m Josh Stroud, your friendly Superaffective AI CEO :) Nice to meet you! I’m based in San Francisco, and I’m excited to work with your business or investment group to build The Next Build Thing :)

Years ago, I founded a company called Sana Health as Technical Co-Founder and CTO. We raised a $25M Series A led by SOS Ventures with participation from Founders Fund (Peter Thiel, PayPal Mafia) and Dreamit Ventures. We built a Medical Device which is approved as a FDA Class I Medical Device in 2026. Sana Health has received 7 patents to date, and is still active today.

I’m an Open Source Software project contributor on projects like Microsoft’s TypeScript compiler in Go, Meta / Facebook’s React, and Speedy Web Compiler, the JavaScript compiler behind Vercel’s Next.js. While I haven’t made pull requests to these projects yet, I plan to submit code updates using Anthropic’s Claude Code by the end of 2026 :)

I’ve worked in AI since 2019, since the birth of the modern Transformer architecture at companies like OpenAI and Google. GPT-2, an early LLM transformer from OpenAI, came out in 2019. When GPT-3 from OpenAI came out in 2021, I started building projects and prototypes with LLMs to understand Generative AI better. I’ve used GPUs in the Cloud to host and run open source AI models, and understand running GPUs and AI workloads in the cloud using Nvidia GPUs for Generative AI Images, Music and Audio, and Video.

Today, I work in Generative AI Video. Superaffective AI has trained an Open Source, Open Weights Generative AI Video model. The model is optimized to run an Nvidia Hopper H100. Developers and AI Transformer Fans can download the model from Hugging Face here. The model can generate videos up to 1 hour in length - Perfect for YouTube, Instagram, and TikTok.

I trained as a software engineer, hardware engineer, and product designer at UC Berkeley Engineering 2010 - 2014. UC Berkeley is the #1 nationally ranked university with over 60 Nobel Laureates and 32,000 undergraduates in 2026 - It’s a big place, and I enjoyed my time at UC Berkeley quite a bit :)

I have 10 years experience as a frontend engineer, backend engineer, and distributed systems engineer, including experience at several startups in Technical Co-Founder and CTO roles, and experience as a software engineer at growth-stage 100-person software startups. I’ve written code in React, JavaScript, and TypeScript, Ruby on Rails and Ruby ERB, Python, Rust, C and C++, Unix and Bash, Ansible, and Docker including using Docker Hub for GPU workloads.

My all-time favorite coding project was building an internet-enabled prosthetic leg in UC Berkeley as part of a graduate level EECS, Mechanical Engineering, and Product Design class. The class was taught by a UC Berkeley Engineering professor, and encouraged us to think across disciplines and combine mechanical engineering, electrical engineering and circuits, software engineering, and product design in a group project. I enjoyed building the software interfaces and circuits in the class, and presented it to the UC Berkeley Engineering community including our project sponsor to rave reviews. Go Bears!

My favorite Cloud Services are Vercel, Cloudflare, and Amazon Web Services (AWS). Vercel is great to build and deliver frontend websites and web apps using React and Next.js with great user experience (UX) and developer experience (DX). I used Cloudflare for 3 years for my image, audio, and video hosting, GPU and AI workload, and database services, and enjoyed using their excellent developer platform. I’ve used Amazon Web Services since 2019 for a number of projects, including building AI and web app prototypes that reached 1,000s of users through LinkedIn and social media marketing. My favorite AWS services are AWS Elastic Compute Cloud (EC2) and AWS Elastic Container Service (ECS) - I’ve used them on a number of projects, and find them useful and reliable.

I like coding, and would happily write code or generate code with an Agentic AI coding tool instead of going to a meeting. One of my all-time favorite startup coding essays is Paul Graham’s “Maker’s Schedule, Manager’s Schedule”, written in July 2009. Paul makes the point that coders need time for thinking time and “deep work”. I’ve found this point to be true over and over again in my career. Thanks, PG!

It’s so nice to meet you on GitHub :) I’d love to hear from you! Drop me a line, and stay in touch.

I can be reached at josh@superaffective.ai - This is my work email, and I check it daily. Open source contributors and maintainers can reach me at this email to discuss projects.

You can also find me on LinkedIn. I have 5k followers on LinkedIn in April 2026. I post daily about Millennial and Gen Z Culture, Startups, and AI and Software Engineering. Feel free to connect or follow me on LinkedIn. I look forward to hearing from you soon!

Until next time, GitHub Fans. 👋 This is your friendly Superaffective AI CEO Josh Stroud, signing off :)