Looking back on my apprenticeship experience!

“You won’t know what you’re *really* capable of unless you try.”

I did it! It’s June 18, 2022, when KadaKareer’s Virtual Apprenticeship Celebration was held. The program itself started from May 9 to June 12, 2022. I am grateful for the KadaKareer Community and its organizers, together with the partner companies of the said program. Everyone was supportive and encouraging! I even won a “Kinareer Award” from them! I specifically worked with DashLabs.ai as one of their apprentices (more on that later!).

I’ve decided to participate in this program to push myself. I opted to undertake it while still in my first year of the Computer Science program. I reasoned that I shouldn’t restrict myself to college classes. I wanted to study more, and I felt Kadakareer’s Virtual Apprenticeship was ideal for me at the moment.

KadaKareer’s Virtual Apprenticeship helps students like me to have experience in solving real-world challenges and problems. With the coaches’ guidance throughout the program, apprentices will be able to navigate through the program. As a Kadet of the community, I joined through their email about VirApp. I originally brushed off the thought of joining, assuming that since I’m still in school, I’m not yet ready. I’m glad I took the chance!

We were given assignments based on the partner company we selected during the course of the four-week program. I worked on creating a chatbot for DashLabs.ai, a partner company. I’ve made the decision to take on the difficulties on my own throughout the entirety of the apprenticeship (there is an option to be in a group as well). They are a company that was founded at the time of our country’s introduction to the Enhanced Community Quarantine. They serve small and medium-sized diagnostic laboratories in emerging countries with end-to-end DIY Lab software, giving them the means to test, register, and manage patients. Their software helps the healthcare workers focus on their work easier, minimizing the paperwork, and helps the patients to have an easily accessible healthcare service through their website.

Before any of the weeks started, we were tasked to read the provided challenge brief, research our chosen company, and attend the virtual apprenticeship onboarding session for more information.

In the first week, the first challenge was to map the different interactions that users can have with the chatbot. During this week, I have studied their provided materials and learned about IX flow (interaction flow) and the common symbols they use for representing ideas. Thankfully, I have learned programming flowcharts before and they are pretty similar making this process easier. I have used Whimsical (a website tool) to help me easily map out the interactions.

I have done the following during this week:

Interacted with their existing messenger chatbot to have a baseline.
Researched through their Facebook page and website FAQs for additional information.
Hand-drawn the flowcharts then transferred to Whimsical.
Finalized the flow chart and then added comments for additional explanation.

The second and third weeks came around and it was about building the prototype of the chatbot using Google DialogFlow. At first, I was not sure and was overwhelmed by everything, especially since it was my first time using DialogFlow! I have only used messenger for building chatbots before. Good thing our coaches were very helpful throughout the whole process. They have taught us a lot and there are provided materials as well.

I have done the following during these weeks:

I have studied my IX flowchart again.
I have learned how to create an agent (the program itself).
I have explored the preset intents and have adjusted them according to my IX flowchart made during the first week.
I have also experimented with entities, actions, and parameters.

For the last week of the challenge, the task was to program exceptions into the chatbot. We needed to take into consideration the instances where a representative of their company (DashLabs.ai) is needed, not the chatbot. This week was the most challenging for me. Our coaches had given us a relaxed version of the task in which we would be needing to outline how we were to use the sentiment score in the interaction flow.

I have done the following during these weeks:

I have enabled sentiment analysis for my chatbot.
I have explored node.js for the first time (more on this later).
I have explored socket.io for the first time (more on this later).
I have tried to do the live agent hand-off but haven’t figured out how to connect it to the messenger integration. *

Snippets of the said live agent (from the video demos I had passed):

Picture 1. The customer is currently typing their request for a live agent.

Picture 2. When the agent turns to the operator interface, a dialog box indicating “Operator requested!” is shown.

Picture 3. The agent is currently typing to further communicate with the customer.

Picture 4. The agent’s response is reflected in the customer interface (this will be shown on the messenger chatbot had I managed to connect this live-agent handoff).

*I have used a great resource for guidance in doing this task. Here is the Github repository: https://github.com/dialogflow/agent-human-handoff-nodejs

My main struggles during the weeks of developing the chatbot are time constraints (due to my studies as well) and the adjustments needed every time I needed to study new concepts. These concepts help me better my chatbot. Nevertheless, I have always enjoyed learning and executing new things, especially sentiment analysis which looked interesting to me the most.

Furthermore, These things are new to me and I’m still learning. The following are the technical things I have learned throughout my apprenticeship journey:

I am grateful for KadaKareer’s Virtual Apprenticeship program, their community, and coaches, and of course the partner companies, (DashLabs.ai for my chosen company!). I’ve learned to be persistent and to allow my desire for knowledge and experience to push me beyond my comfort zone thanks to this entire journey. It is okay to feel lost because everyone has to start somewhere and it is better than not trying at all. As of now, I am developing the ability to progressively teach myself. I would really like to do more work in the artificial intelligence and machine learning fields in the future. I’m excited to gain new knowledge and broaden my experience!

Here is my certificate of completion!