Enhance developer experience by allowing easier API discovery and saving developers’s time on API integrations.
Alfred is an AI developer portal tool, and one might even call it an AI as a service product. With Alfred as a Service, users can take our fully featured AI assistant, and embed it into their own dev portal!
There is no need to use Treblle platform or any other Treblle product for this.
Integration is very easy, 1 line of javascript code + 1 line of HTML. See Treblle documentation for more information.
This instantly gives internal and/or external API consumers an AI-assistant that answers questions and automatically generates a wide range of data that might be needed.
Think of Alfred as a high-tech librarian. Just as a librarian assistant helps you find books, Alfred helps developers find and use tools and data they need for building apps. It’s like having a helper who understands exactly what you’re looking for and brings it to you instantly.
- Mobile engineering teams can generate models and code to make requests.
- Partners can build SDKs and integrations to your API faster.
- Users can get answers to help them understand your API more quickly and easily.
Alfred AI announcement video
The developer experience for your internal teams is immediately improved. And most importantly, the quality of your APIs are available to everyone, which will in turn put positive pressure on your teams to build higher-quality APIs.
All of this leads to better business contributions from the API engineering organization. APIs for the win, again.
Alfred AI’s Character Makeover - How We Have Improved Alfred
It has been 10 months since we launched Alfred, our AI-powered assistant. And as it goes with the software world, a lot has changed since then.
We initially built Alfred to solve a basic problem: create a tool to automatically generate code for SwiftUI models based on the API documentation.
We wanted to save users time and make their lives easier. Since the Treblle platform automatically generates docs based on an API implementation (via our SDKs), the creation of Alfred was a way to do this.
That initial solution turned into an expanded product where users land on our auto-generated API documentation page in the Treblle platform and then interact with AI to do things such as generate code samples and models for whatever development language they need (beyond just SwiftUI) as well as tests.
The main 3 tasks we wanted Alfred to tackle were:
- Automate API Integrations
- Upgraded Developer Experience
- Enable Doc-Driven Development
The Makeover
While early comparisons of our AI solution naturally went to Batman who is the “worlds greatest detective” in the DC universe, on deeper reflection, we knew it was much more akin to Alfred, Batman’s beloved, loyal, and steadfast butler.
The next thing that changed was Alfred’s logo:
Sure, we liked the stern-looking butler face, but it wasn’t reflective of who we are as a company and where we’re going. While the name continues to reflect how we feel about our AI, change was in order.
The next change came with the enhancement of allowing Alfred to answer more complex questions about your API Docs. For example:
- Are these docs using Swagger? Nope, it’s OpenAPI 3.0.x.
- Can you show me how to make a POST request to an endpoint? Sure! And the instructions follow.
- Is there a PUT request defined in the docs? Depends on the docs, but Alfred will tell you the correct answer every time.
- Can you tell me the quality of my API? Yes I can.
- What is the API Key to get started with this API? Here you go.
- Are you Batman? Nope, I’m just an AI assistant.
- Can you sell me a car? Sorry, I can just give you information related to APIs.
We then integrated our API Insights governance tool directly into Alfred. API Insights is our solution to allow developers to quickly see and understand the quality of their APIs based on the API specifications PLUS an actual request to their API (so it's more than just lining).
Essentially we take the existing specification/docs as well as request/response data from the API call and run them through a series of tests to ensure that you are following industry standards and best practices across API design, performance, and security.
Following API Insights, we launched a MacOS-native API client called Aspen. Aspen is a lightweight, login-free client that runs and stores data locally so users don’t have to worry about sending data to the cloud.
Not only can users make an API request, we brought in Alfred to provide users with the same AI experience but based on an API request rather than the full API documentation: users can generate data models, integration code, and more in any programming language.
With this functionality, users can seamlessly move from seeing the request flow and understanding an API response to integrating that API in seconds.
Culmination: Embed Alfred into your own Developer Portal
Flash forward to the last few weeks. Some of our customers have been talking to us about their dev portals and how they would like to make tasks easier for their API consumers and dev partners.
Some companies were even budgeting for a team to start building developer portal tools - taking time away from their core offerings. We were like, wait – we have a solution for this! You can embed Alfred directly in your dev portal and get instant AI-assistance with almost zero work on your side.
With seamless integration and instant AI assistance, Alfred empowers your API consumers and development teams to streamline their workflow, enhance productivity, and focus on what they do best—innovating. Don't miss out on transforming your developer portal into a high-performance hub.
Alfred AI
The assistant for modern developer portals.
Alfred is an AI developer portal tool that helps with API Discovery, Adoption and Integration.
