Skip to main content

6 posts tagged with "developers"

View All Tags

· One min read
Josh Twist

Today we're excited to announce one of the first cool features resulting from Zuplo's bet on the OpenAPI standard: rapid mocking of APIs using examples inside an OpenAPI document - it's just what every

Sometimes it's better to show than tell, so in that spirit, check out the 2 minute video below that shows you the end-to-end experience.

With this powerful new capability, developers can effortlessly generate mock endpoints that mimic the behavior of real APIs, speeding up development cycles and enabling more efficient testing. Combined with our sleep policy you can even mix in some latency for added realism - just what every mockstar frontend developer needs!


To get started, create a new project in Zuplo at (or use an existing project if you have one). Hopefully you have an OpenAPI document (or you can get ChatGPT to generate one for you) with some juicy examples in it - I'm going to use the Postman OpenAPI doc from

Let's import your OpenAPI doc.

Import OpenAPI

Don't forget to save your changes. Save your changes (CMD+S).

· 3 min read
Josh Twist

I recently had the pleasure of chatting with Ben Garvey, the co-founder and CTO of Common Paper. Our discussion covers their innovative open-source platform for contracts and how they are revolutionizing the way businesses handle contract management and how they used Zuplo to accelerate the launch of their API.

Unleashing the Power of Open Source for Contracts

In the chat (video below) Ben explains that Common Paper is on a mission to simplify and accelerate the contract creation and negotiation process. Their platform offers a comprehensive range of standard agreements that can be easily downloaded from their website. From NDAs to CSAs, DPAs to partnership agreements, Common Paper provides a vast library of open-source contracts. This approach draws inspiration from the world of open-source software, where licenses like Apache 2 and GPL are readily available for developers without the need for legal consultation.

By leveraging Common Paper's platform, startups and businesses can say goodbye to the archaic practice of emailing Word documents back and forth during the contract negotiation phase. Instead, the platform enables users to create, negotiate, and sign contracts directly within the app. This streamlined workflow not only saves time but also reduces costs, empowering sales teams to close deals faster.

Deploying Zuplo to focus on what matters and move fast

During our conversation, Ben highlights the benefits of Zuplo - acting as a gateway for Common Paper's API, with essential capabilities including API authentication, rate limiting and developer documentation. With a small team at Common Paper, leveraging Zuplo allowed their engineers to focus on product development, leaving the complex aspects of API management to the experts.

One interesting aspect of Common Paper's integration with Zuplo is the generation of API keys being embedded into their own dashboard. By utilizing Zuplo's API key service, Common Paper simplifies the process of managing and securing their API keys.

Ben notes the excellent service he has received from the team at Zuplo and is a particular fan of how fast Zuplo is - particularly our near instantaneous deployment time; aligning with Common Paper's need for quick iteration and shipping fast. The combination of Zuplo's reliable infrastructure, extensive documentation, and responsive support made it an ideal choice for Common Paper's API management requirements.

If you're looking to simplify your contract management checkout Common Paper, and if you're interested in accelerating the launch and simplifying the operations of your API - try Zuplo for free at

· 6 min read
Nate Totten

If your business provides and API to your customers and partners (which it almost always should!) it is important to build your API in a way that helps developers be successful. While there are many elements to API design, this post will discuss the choice between GraphQL and REST. GraphQL is often an excellent choice for internal APIs but might not always be the right fit for customer-facing or partner APIs due to its complexity. This post will discuss some of the tradeoffs between GraphQL and REST as well as offer some ideas on how this doesn’t have to be an OR decision, but rather how you can relatively easily support both GraphQL and REST.

GraphQL: Powerful, Performant, and Complex

GraphQL, a query language for APIs, was developed by Facebook in 2012 to address the limitations of traditional REST APIs. A GraphQL API can make your frontend more performant by allowing fine-grain queries that allow fetching exactly the data needed for your app. This makes a lot of sense for certain apps (especially super complex one’s like Facebook). Some additional benefits of GraphQL are:

  1. Flexibility: GraphQL allows clients to request the exact data they need, without over-fetching or under-fetching. This enables faster and more efficient communication between services.
  2. Strong Typing: GraphQL's type system enables API developers to define the types of data that can be requested, making it easier to maintain and debug applications.
  3. Schema Introspection: GraphQL provides built-in tools for exploring and understanding the API, making it easier for developers to work with and build upon.

Despite its numerous benefits, GraphQL's complexity can be a challenge when it comes to external APIs shared with customers or partners. Here's why:

  1. Steeper Learning Curve: GraphQL's query language and type system can be daunting for developers who are new to the technology. This can lead to increased onboarding time and difficulty when it comes to integrating with external systems.
  2. Security Concerns: GraphQL's flexibility can also lead to potential security issues, such as exposing sensitive data or allowing clients to perform costly operations that could impact server performance. This requires careful implementation of access controls and query validation.
  3. Resource Intensive: The flexibility of GraphQL means that queries can becoming extremely complex and taxing on your backend. For example, a request might end up requesting data from multiple tables in your database over a large timespan. You must be extremely careful when implementing your API to ensure customers are generating queries that tax your systems.

REST: A Simpler Choice for External APIs

Considering the challenges posed by GraphQL's complexity, REST is often a better choice for APIs shared with customers or partners. Here's why:

  1. Simplicity: REST is based on standard HTTP verbs (GET, POST, PUT, DELETE) and status codes, making it intuitive and easy to use for developers who are already familiar with HTTP.
  2. Compatibility: REST APIs can be easily consumed by various clients, including web browsers, mobile apps, and other servers, without the need for specialized libraries or tools.
  3. Caching: REST APIs support built-in HTTP caching mechanisms that are easy to implement and will improve performance and reduces server load.
  4. Mature Ecosystem: REST has been the dominant API design style for years, resulting in a mature ecosystem with extensive documentation, tooling, and libraries. Everybody developer, on any platform, with any language can use a REST API.

Making the Choice (or not)

So say you really want to use GraphQL (or maybe you already built a GraphQL API), but now you want to share your API with customers or partners. For the vast majority of scenarios, REST is a better choice for an external API. Developers will onboard faster, you will spend less time supporting your external API program - it’s just easier. Fortunately, you don’t have to choose. There a several great libraries out there that make it easy to proxy a REST API to a GraphQL backend.

One such library is called SOFA. This is a Javascript library that is compatible with Node, Deno, and standard fetch APIs. It doesn’t have a dependency on any specific framework so it can basically be plugged into any web server.

Building a REST to GraphQL Proxy

In this next section, we’ll explore how to build expose a GraphQL API with REST using SOFA. For this post, we’ll use expressjs, but in an upcoming post I’ll go through how this can be accomplished on Zuplo without having to worry about any infrastructure and how you can deploy to our 250+ datacenters at the edge instantly.

You can see the full source of this example on Github, but lets go through the important parts below. First, install create a new project and install the dependencies.

npm init -y
npm install express sofa-api @graphql-tools/schema

For this sample, we’ll use ESM modules so open your package.json and add the property "type": "module".

Next, create a index.mjs (make sure to use mjs for ESM) file with a simple express server. Add the sofa module to the /api route.

import { useSofa } from "sofa-api";
import { makeExecutableSchema } from "@graphql-tools/schema";
import express from "express";
import { resolvers } from "./resolvers.mjs";
import { typeDefs } from "./schema.mjs";

const app = express();

const schema = makeExecutableSchema({

basePath: "/api",

app.listen(3000, () => {
console.log(`Example app listening on port 3000`);

Next, add your schema. There are several ways to do this, but to keep it simple here it will just be a string. Create the file schema.mjs and add the schema. The full file can be found in the example.

export const typeDefs = `
type Pizza {
dough: String!
toppings: [String!]

type Salad {
ingredients: [String!]!

Finally, create a mock resolver in resolver.mjs that just returns data from fixed collections. In a real app this would be your GraphQL resolver.

export const resolvers = {
Query: {
me() {
return UsersCollection.get(1);
user(_, { id }) {
return UsersCollection.get(id);

Run npm start in the project directory and query the server. Try navigating to https://localhost:3000/api/users or using your favorite API client to fetch the list of users.

REST API users response

Wrapping Up

While GraphQL is a powerful and flexible choice for building internal APIs, its complexity can make it less suitable for customer-facing or partner APIs. REST, with its simplicity and ease of use, is often a better choice for external APIs. Ultimately, the decision between GraphQL and REST should be based on your specific use case, the needs of your developers, and the requirements of your customers or partners. Or if you find yourself in a position of not wanting to choose, you can use a few great open source libraries to expose your existing GraphQL API to your external users as a REST API.

· 5 min read
Josh Twist

Over the past few years, the field of natural language processing has seen rapid advancements, with GPT-based models like Chat-GPT leading the charge. It's an exciting time with massive productivity gains ahead, especially for software engineers.

One area of software development that is going to grow exponentially important in a world boosted by AI productivity is APIs. APIs are how AI will actually get things done and interact with real-world services.

Proof of this came recently with the announcement of Chat-GPT plugins. This is a way for OpenAI to connect with real-world services via APIs. And, amazingly, all OpenAI need to understand how your API works in an OpenAPI document. Wow, we knew OpenAPI (so confusing, the similarity in names) was going to be important but...

As a company that is all about APIs all the time, we couldn't be more excited. We're the only gateway natively powered by OpenAPI which makes us the fastest way to deploy a Chat-GPT plugin. In this post, I'll show you how (also available in video)

For the purposes of this demo, we'll use a simple example API

1/ Installing a plugin to Chat-GPT

First, you need access to the Chat-GPT plugin which is currently in alpha. I found it hard to learn how to do this so I'll share in detail here - but you will need access.

Login to Chat-GPT at and make sure you have the Plugins option selected

Chat GPT Plugin

Then, to add your own plugin, go to the plugin store

Plugin Store

In the plugin store, choose Develop your own plugin at the bottom

Develop your own plugin

This will open a prompt, asking for your domain - where OpenAI will find your manifest file.

Enter URL

The manifest file is a document that tells OpenAI more about your API and where it will find your OpenAPI document. It has to be at the same location on your domain With Zuplo, we can easily create and host a manifest - let's switch over to Zuplo and get everything ready, including our manifest and OpenAPI document.

2/ Creating the manifest file

Create a free account on Zuplo at and create a new project. In your project - we'll need to add our first route for the manifest.

Click on routes.oas.json to select your routes file (this is actually an OpenAPI file under the hood) and click Add route.

Set the following properties on your new route

  • Summary: OpenAI Manifest File
  • Path: /.well-known/ai-plugin.json

And press CMD+S to save your changes. Next, we'll need to make the handler for this route a custom function. In the Request Handler section, choose Function and use the ellipsis ... to create a new Module called manifest.

Create manifest.ts

This will open your new manifest.ts file in the editor. Paste the following code, replacing the original template:

import { ZuploContext, ZuploRequest } from "@zuplo/runtime";

const manifest = {
schema_version: "v1",
name_for_human: "Pro Club Tennis",
name_for_model: "pro_club_tennis_score",
"Plugin getting the current players and score for Tennis on our 6 courts",
"Plugin getting the current players and score for Tennis on our 6 courts",
auth: {
type: "none",
api: {
type: "openapi",
url: "/openapi",
is_user_authenticated: false,
contact_email: "",
legal_info_url: "",

export default async function (request: ZuploRequest, context: ZuploContext) {
// Zuplo will automatically serialize to JSON and add the content-type header
// if you return an object like this in a function.
return manifest;

A few things to note about this manifest, we are not using any auth for our API - it's public (you can add auth protection with zuplo if you like). Also, note the relative link to our OpenAPI file at /openapi. That's the next route we'll add. Save your changes - your manifest file is now ready.

3/ Serving the OpenAPI file

Since our routes.oas.json file IS an OpenAPI file we just have to tell Zuplo that we'd like to serve it, and on what path.

Add a new Route and set the path to /openapi. Choose OpenAP Spec for your route handler, it should look like this:

OpenAPI route

This will now serve a sanitized (x-zuplo stuff and x-internal operations removed) from the path /openapi. Neat! The very routing table powering your gateway and documentation is now available as an endpoint (and visible to OpenAI).

4/ The actual API part

If you have an OpenAPI file for your API already, you can just import to Zuplo and most likely you are off to the races. To do this, go to the OpenAPI tab and hit Import OpenAPI.

Import OpenAPI

Your routes should have been imported and you should be good to go.

If you don't have an API with OpenAPI handy, you can easily use Zuplo to create one. Watch this video to see how I created the demo API using Zuplo + ChatGPT in under 5 minutes.

5/ Setup the ChatGPT plugin

Now it's time to go back to chat GPT to setup the plugin, but before we do go the Getting Started page to get your gateways URL.

Getting started in zuplo

Use the copy button next to your gateway's URL ("Your API gateway is live at").

Now head back to ChatGPT and paste the URL into that dialog asking for your domain

Enter your domain

And click Find manifest file.

With a bit of luck, you'll see the next 'Found Plugin' screen with no errors. Otherwise errors in your manifest are typically well explained and easy to correct.

Found plugin

If so, click Install for me and keep going through the many dialogs until you see your plugin in the chat gpt main screen. Then ask, 'What are the scores on court 3?'

Et voila! A working plugin

Working plugin

Checkout the video version of this content:

· 10 min read
Josh Twist

In 'The three pillars of an API program' we outline that one of the critical pillars of launching an API is protection - or, more specifically, rate limiting.

API rate limiting is a technique used to control the number of requests that can be made to an API within a specified time frame. This helps maintain the stability and performance of the API by preventing server overload, protecting against DDoS attacks, and ensuring fair usage among consumers. Additionally, many businesses use rate limits as a product feature (e.g. premium customers get higher rate limits than free, for example - see Twitter).

There are infinite articles about different rate limiting algorithms - token bucket vs sliding window, yada yada. Mostly these articles are more interesting academically than they are practical; there are more pressing concerns that must be closed before thinking about the nuances of different algorithms.

That’s why we’ve put this guide together for you, based on our experiences working with many customers on rate-limiting challenges at zuplo, including building a high-performance distributed rate-limiting capability. How distributed? Try 1000s of instances per customer running in 250+ data centers around the world.

This article is also available in video

We’ll cover

  • a canonical rate-limiting implementation for an HTTP API
  • secret limits or public?
  • how should rate limiting be applied? globally, by IP, user, or custom functions?
  • rate-limiting in distributed systems
  • the latency/accuracy tradeoff
  • observability - what do I need?
  • how do I configure my rate limits in the first place?

This article explores these items from the viewpoint of the API owner and how to offer the best experience for your API consumers.

A canonical approach to rate-limiting for an HTTP API

This part is relatively simple but is often done incorrectly. We recommend returning a 429 status code which is described on MDN as “indicates the user has sent too many requests in a given amount of time (’rate limiting’)” - so perfect.

We recommend using the Problem Details format as the body of your response. For example:

"type": "",
"title": "Too Many Requests",
"detail": "You made too many requests to the endpoint, try again later",
"instance": "/account/12345/msgs/abc",
"trace": {
"requestId": "4d54e4ee-c003-4d75-aba9-e09a6d707b08"

Depending on the decision you make in the next section, you might choose to include a Retry-After header with the number of seconds the user should wait before trying again.

Retry-After: 3600

Do you keep limit thresholds and timings a secret?

Some rate-limiting implementations return information to the user that tells them how long they should wait until trying their request again. This can be very useful but also presents some risk. If a malicious user is effectively trying to maximize the amount of resources they can consume without hitting the rate limit, they can use this information to infer your rate-limiting policy. For this reason, many businesses choose not to disclose this and keep rate-limit thresholds and timings private. This is common for public, anonymous APIs but also for APIs that require authentication, but it’s easy to get an access key. For example, a free tier of a service where I can use an e-mail address to get access to the API. The gate to entry isn’t effortless, but it isn’t hard either.

However, if you have a more formal arrangement with your API consumer; maybe they are a paying customer, then it is often better to share details of the rate limits you’re imposing on them. So we find most B2B APIs that require a contract are liberal in disclosing their rate-limits, while free APIs keep their cards closer to their chest. Based on this, you may choose not to share a Retry-After header (and we made this an easy option in the Zuplo rate limiting policy).

You may even choose to have a variable policy - where your free customers don’t get a header, but your paying customers do.

How to apply rate limiting (AKA why IP or global rate limiting is a bad idea)

So you’ve been convinced that you need to think about rate-limiting protection 👏 - the next step is to think about how to apply that rate limit.

The most obvious approach might be to apply a global rate limit to all incoming requests. That will help make sure your backend doesn’t go down. But remember your goals, you want to protect your resources and ensure a great customer experience by not allowing one customer to impact others (by taking you down or abusing their API access).

It may still be valuable to have a global rate limit, but you will need something more fine-grained to prevent any noisy neighbor issues. At this point, folks might reach for IP-based rate-limiting. However, in some scenarios, this can actually increase the chance that one customer negatively impacts the experience of another.



If you have two API consumers hosting at a major cloud provider, it’s possible their outbound calls share an IP address. For individual users calling your API from their device, they might be co-working in a WeWork and sharing WiFi, and most likely an IP address. In these cases, rate-limiting by IP can actually create a noisy neighbor problem, as one customer causes the other to be blocked when they hit your rate limit. This is surprisingly common.

The best approach is to rate limit by the user, typically leveraging an authentication mechanism like a JWT or API Key. This is the fairest approach. In Zuplo, when we do user rate-limiting, we unpack the JWT (or API key) and extract the sub or ‘subject’. So even if the same user has two JWTs, they’ll both count toward the same rate limit total.

The most advanced approach allows you to segment users into different buckets. For example, maybe free users get 10 RPS (requests per second), while premium users get 1M RPS. A good example of somebody who does this is GitHub - they even allow unauthenticated access to their API, but it is tightly throttled.

This is user-specific rate-limiting (we call it dynamic rate-limiting in Zuplo).

Rate-limiting in distributed systems

Implementing a rate-limit algorithm or using an open-source package is relatively trivial if you have a simple setup with a single server (or process) handling the rate-limit counting. You can probably just keep the count of requests in memory for maximum performance and the lowest latency. However, the minute you scale up your system (for high availability or higher load), you have a whole new set of problems.


You cannot use an in-process counter because your load balancer might send the next request for a given user to a different machine, allowing them to exceed the rate-limit many-fold (by as many machines/processes as you have). One way around this is to have a sticky session, so all traffic from a given identity goes to the same server. Building stateful services like this is generally a terrible idea, and rate-limiting is no longer your most pressing concern. Go fix that, then come back.

So most folks will need to think about an independent service to synchronize counts. Many folks use Redis for this. The problem now is that you are adding latency - often a meaningful amount - to your API.


This becomes a particular problem if your service is highly distributed (or edge-based, like Zuplo) and in many data centers around the world. Do you have a single Redis instance in one location? That will add huge latency for folks calling servers far away from that location. The correct answer is to have multiple services around the world with some form of synchronization. This is non-trivial to setup, deploy and manage — but it is the world we live in today as developers.

At Zuplo we’ve spent a lot of time optimizing that latency with layers of caches and advanced replication. We call this strict mode, and when configured, we will not call the backend (API) until we know for sure this user is not rate-limited. However, we also have an async mode if you would prefer a different tradeoff…

The latency/accuracy tradeoff

In the design above, we are trading away performance (latency) in favor of accuracy. For many businesses, this might not be the right decision. Every call to your API will have added latency even though your API consumer has done nothing wrong. Instead, you may choose a design that gives the user the benefit of the doubt and allows the request to run in parallel with the rate limit check.

In Zuplo, we call this async mode.

We implement a local cache (a datacenter-level cache, in-memory cache, or combination thereof). Now the latency added to each request is minimal. If the rate-limit check comes back negative the user will still get a 429 Too Many Requests (and we will update the local cache with an appropriate timeout to match the Retry-After).


The downside to this approach is your backend may receive a little more traffic than was strictly defined. In most cases, this is probably the right call to offer the best experience for your API overall.

Observability - what’s the minimum you need

So far, we’ve only looked at how you might design and implement rate-limiting. We have not considered some of the important non-functional requirements like observability and analytics. Even better, how do you know where to set your rate limits?

A minimum bar for reporting on your rate-limiting includes

  • See a count of rate-limited responses
  • Being able to break this down by user (or IP, etc)

With these, you can identify users that are getting rate limited, how often, and reach out to them if necessary to discuss remediation (or punishment 😄).

To go further, it’s also good to be able to see the overall RPS (request per second) of your system and users as this will help with the next, and final section:

How do I configure my rate limits in the first place?

Setting rate limits is hard? How do you know when your system will break? Most businesses don’t get to stress test their system to the actual breaking point to set these limits, so the following approaches and heuristics can be helpful:

  1. Test-To-Failure Data - if you have had the benefit of a formal Test-To-Failure, then you should have some data on your max capacity and can use this to inform your rate limiting decisions. Maybe you didn’t proactively set this up but were unexpectedly tested in a real-world scenario. This data can be valuable to understand your breaking point. You might divide your overall capacity by your number of customers, plus some margin
  2. Observation before enforcement - before you enforce rate limiting, you can have a period of observation and look at the max RPS of your different customer profiles. You might then typically set the rate limits at some reasonable margin higher than that max, assuming that any meaningful spike beyond this is probably not intentional. Zuplo’s Rate Limiting and RPS reports can be useful for this.
  3. Business / Pricing - rate-limiting might be part of your overall business strategy, and you may charge customers for higher rate limits. In this case, conversations with your customers and prospects (combined with observation-before-enforcement) are usually a great source of input for how to set these rate limits.

Hopefully, you found this useful. As always, if you don’t want to build or manage any of this yourself, you can always turn to solutions like Zuplo that offer a fully managed, distributed, and programmable rate-limiting capability - try it for free at

· One min read
Josh Twist