Source URL: https://simonwillison.net/2025/May/13/launching-chatgpt-images/#atom-everything
Source: Simon Willison’s Weblog
Title: Building, launching, and scaling ChatGPT Images
Feedly Summary: Building, launching, and scaling ChatGPT Images
Gergely Orosz landed a fantastic deep dive interview with OpenAI’s Sulman Choudhry (head of engineering, ChatGPT) and Srinivas Narayanan (VP of engineering, OpenAI) to talk about the launch back in March of ChatGPT images – their new image generation mode built on top of multi-modal GPT-4o.
The feature kept on having new viral spikes, including one that added one million new users in a single hour. They signed up 100 million new users in the first week after the feature’s launch.
When this vertical growth spike started, most of our engineering teams didn’t believe it. They assumed there must be something wrong with the metrics.
Under the hood the infrastructure is mostly Python and FastAPI! I hope they’re sponsoring those projects (and Starlette, which it FastAPI under the hood.)
They’re also using some C, and Temporal as a workflow engine. They addressed the early scaling challenge by adding an asynchronous queue to defer the load for their free users (resulting in longer generation times) at peak demand.
There are plenty more details tucked away behind the firewall, including an exclusive I’ve not been able to find anywhere else: OpenAI’s core engineering principles.
Ship relentlessly – move quickly and continuously improve, without waiting for perfect conditions
Own the outcome – take full responsibility for products, end-to-end
Follow through – finish what is started and ensure the work lands fully
I tried getting o4-mini-high to track down a copy of those principles online and was delighted to see it either leak or hallucinate the URL to OpenAI’s internal engineering handbook!
Gergely has a whole series of posts like this called Real World Engineering Challenges, including another one on ChatGPT a year ago.
Via @GergelyOrosz
Tags: chatgpt, generative-ai, gergely-orosz, openai, scaling, ai, llms, python
AI Summary and Description: Yes
Summary: The text discusses the launch and scaling of ChatGPT’s new image generation feature, highlighting the technical infrastructure used, user growth, and insights into OpenAI’s engineering principles. It offers valuable information for AI professionals, particularly in understanding backend scaling challenges and workflows within AI applications.
Detailed Description: The content provides a comprehensive overview of the recent developments surrounding ChatGPT images, focusing on a conversation with key OpenAI engineering figures.
Key points include:
– **Successful Launch**: ChatGPT images were introduced in March, leading to a viral user adoption, with one million new sign-ups in just one hour and reaching 100 million in the first week.
– **Scaling Challenges**: Initially, there was skepticism regarding the growth metrics within engineering teams. To tackle early scaling issues, OpenAI implemented an asynchronous queue system for load management, which resulted in longer generation times for free users during peak hours.
– **Technical Infrastructure**: The infrastructure primarily relies on Python and FastAPI, showcasing the tech stack critical to supporting such a scalable application. Additional technologies mentioned include C and Temporal for workflow management.
– **Engineering Principles**: Key engineering philosophies described are:
– **Ship relentlessly**: Emphasizing the importance of rapid deployment and continuous improvement without waiting for ideal conditions.
– **Own the outcome**: Engineers are accountable for their products throughout their lifecycle.
– **Follow through**: Ensuring that projects are completed fully and transitions are smooth.
– **Insights and Observations**: The piece subtly hints at the potential existence of a detailed internal engineering handbook at OpenAI, underscoring the operational transparency and thoroughness valued in their development processes.
Such insights into the operational dynamics at OpenAI are particularly significant for security, compliance, and infrastructure professionals looking to understand the technical challenges and principles that guide the rapid innovation in AI technologies.
The text also emphasizes the implications of scaling AI tools efficiently while maintaining a solid engineering foundation, which is crucial for security and privacy considerations as these technologies grow.