Hello there! I´m a Options Engineer working with Collaboration and Buyer Expertise options, and one factor I’ve realized is “one measurement doesn’t match all.” To ship essentially the most impactful enterprise outcomes, clients want digital options tailor-made to their distinctive wants. Fortunately we’re surrounded by SaaS functions and highly effective platforms that present open APIs obtainable to anybody with the abilities to make use of them. Briefly, in the event you provide the motivation, the instruments are there to embrace automation as the muse for constructing customized use instances. That is how one can actually differentiate within the market, and produce worth to our clients.
My curiosity in leveraging automation to combine functions and optimize workflows was the set off that began my journey into community programmability. I knew I needed to begin by connecting with a group of like-minded individuals who may assist me get began. That’s when I discovered Cisco DevNet.
Getting began with a DevNet Studying Monitor
My first step was to be taught the fundamentals of APIs – how they work, what they’re used for, the way to take a look at them, and the way to combine them into different functions. RESTful APIs are extensively utilized in net growth, permitting functions to work together with one another and trade information in a regular and scalable method. They supply a versatile and environment friendly method for various methods to speak and combine with one another.
To get acquainted with APIs I began training with Studying Labs obtainable within the Cisco DevNet developer portal. As a result of I’m a Options Engineer working with Collaboration options, I select the Webex – Begin Now – Studying Monitor. Comply with this hyperlink to entry this studying observe and plenty of others that may get you began. This Studying Monitor guided me to put in all of the required functions in my PC to begin working with APIs and coding. It additionally gave me the instruments I wanted to begin doing my first assessments with coding and APIs.
As soon as I had some fundamentals underneath my belt, my subsequent step was to create a use case and begin a “actual” programmability mission. I made a decision to combine Webex with ChatGPT utilizing the OpenAI APIs. The primary constructing blocks to implement this mission are:
- Webex Bot
- OpenAI APIs
- Middleware that runs the bot logic
After I realized the way to create a Webex bot, my subsequent step was to get acquainted with the OpenAI APIs. Anybody can create a free OpenAI account. (On the time of scripting this weblog, you get a free credit score of $5.00 to make use of, which is sufficient to do some experimentation with their APIs.)
Getting the fundamentals of Python
Subsequent up, I wanted to create my bot logic. Speaking with my mentor, Julio Gomez, he shared with me a mission that he did for the same use case, and I used to be in a position to reuse his Phyton code. I had realized the fundamentals of Phyton and the way it works. Thus, I used to be in a position to adapt the Phyton code to make use of with my Webex bot and the OpenAI API that I had recognized as essentially the most applicable for growing my Q&A bot based mostly within the ChatGPT consumer expertise.
To make it simpler to adapt the code, I divided it into constructing blocks based mostly on the totally different features: obtain a textual content message from the Webex Bot; course of the message and ship it to OpenAI APIs; obtain the reply from the OpenAI; and at last ship it again to the Webex Bot.
I confronted one other problem
I used to be working the bot logic on my PC and I needed to expose it to the web. The Webex cloud platform requires your bot to be reachable from a publicly accessible endpoint. Utilizing Webhooks, the Webex bot sends notifications as occasions to the endpoint that subscribed to the webwook, for each interplay between you and the bot.
For that objective, I created an inbound tunnel between my native PC and the web in order that the port your bot is listening on turns into publicly accessible. Ngrok is a well-liked instrument I used to create these tunnels. The Ngrok service creates a URL that can be utilized to entry your utility working in your native PC. This URL is then used to create a Webhook in your Webex bot to ship notifications when somebody interacts with it.
After I reached the purpose the place I had a working utility, I used to be prepared to begin my preliminary testing. I needed to wonderful tune the code to acquire the specified consequence. In truth, it was as straight ahead as altering just a few parameters within the JSON payload, associated to the OpenAI APIs. Right here is the documentation of the OpenAI API that I used:
Lastly, I used to be in a position to create a Q&A bot in Webex leveraging the LLM (Massive Language Mannequin) developed by OpenAI. The top customers are then in a position to ask inquiries to ChatGPT from the Webex App, as in the event that they have been chatting within the native ChatGPT consumer interface.
The diagram under is a high-level structure of what I constructed. You’ll be able to see my mission on this GitHub repository.
APIs have reworked the way in which we customise experiences and combine functions. I’m not a software program developer, however I used to be in a position to obtain my objective, which was to construct a Q&A Bot based mostly on new applied sciences like ChatGPT. For me, working with APIs is like enjoying with Lego blocks. It’s enjoyable!
My journey in programming continues
Then one other thought got here to my thoughts – Would I have the ability to create the bot with little or no code, and even no code? I put myself into one other problem, which was to make use of an orchestration platform that leverages a low code / no code framework to create workflows and automation utilizing APIs.
In truth, at Cisco now we have our very personal Low Code orchestration platform. It’s known as Webex Join and is without doubt one of the most acknowledged Communications Platform as a Service (CPaaS) out there.
Webex Join is a platform that empowers the creation of good, proactive, and customized interactions throughout digital channels comparable to WhatsApp, Apple Messages for Enterprise, Google Enterprise Messages, Fb Messenger, Instagram, SMS, and so on. It orchestrates buyer journeys end-to-end by automating enterprise processes.
I opened the drawing canvas of the applying known as Circulate Designer and began constructing my Bot logic. It was so simple as dragging and dropping bins, connecting them with arrows, and following the method. Once more, like enjoying with Legos.
Utilizing this platform is way less complicated, as a result of I didn’t should take care of programming languages and coding for my use case. I simply must grasp the functions I needed to attach, and learn to use their APIs.
In a matter of some hours I had my bot up and working. This time I built-in WhatsApp with ChatGPT. What took me just a few days with Phyton was diminished to some hours as a result of I used to be coping with APIs, webhooks, and parsing JSON payload information. Webex Join (CPaaS) did its “magic” and executed the code behind the scenes which simplified the entire mission.
Right here is the ultimate circulation:
Low code platforms like CPaaS can unlock the Artwork of Potential, with APIs to create pleasant consumer experiences.
I hope this was useful. And maybe even made you curious to begin your personal journey into programmability. In that case, Cisco DevNet is a good place to begin.
Good luck!!
Associated assets
Join Cisco U. | Be part of the Cisco Studying Community.
Comply with Cisco Studying & Certifications
Twitter | Fb | LinkedIn | Instagram | YouTube
Use #CiscoU and #CiscoCert to affix the dialog.
Learn subsequent:
Share:
