What It Means To Go To Prod

Lately, I am feeling more and more confident about Portkey solving *real* production challenges for our users.

Of course, developer love and revenue increase are important parameters, but there's something else just as high signal:

The platform modularity that customers demand from Portkey, and how fast we serve it without adding additional complexity.

Case in Point: Need for a JSON Mode

One of our users recently requested the ability to configure prompt variables en masse with a JSON mode.

Pretty unique request, I'd say. How many companies are going to have 10/20 variables in a single prompt? And how many of them are going to profusely test the prompt until it becomes too cumbersome for them?

But then I realised, over the last months, this same theme has repeated for multiple of our features: Logs, Playground, Configs, and more. Every time, a customer who is deep in prod ends up requesting for a JSON mode on Portkey so that they can configure the feature to their exact workflow.

What Do These In-Production Workflows Look Like?

Admittedly, *only* the people who are at scale have seen and built them. With Portkey, we are really excited to build the exactly-fitting tooling for these developers.

So, let me propose a new metric for measuring the production-readiness of your product - it is directly correlated to the number of JSON views you need for each feature 😆

Ayush and I are going to be in New York this 24th to shed more light on what these production workflows look like and what everyone else can learn from them.

We are bringing together world-class practitioners & leaders who have successfully built and scaled LLM-based apps to production.

If you are in NYC, we would love to catch up at the first LLMs in Prod event on the east coast.