Under the coding hood: Angelhack Kochi Hackathon 2018
It is that time of the year when all the tech enthusiasts of Kochi get together and code their way to glory — all under 24 hours. AngelHack…
---
Under the coding hood: Angelhack Kochi Hackathon 2018
It is that time of the year when all the tech enthusiasts of Kochi get together and code their way to glory — all under 24 hours. AngelHack Kochi was on July 28–29 and boy, there were many who were up for the challenge.Introduction
On the Saturday morning, Nikita Uday, Akshay Balakrishnan and I took the metro to Maker Village at Kerala Technology Innovation Zone, the venue of the hackathon. When we reached, there was a long queue of participants at the registration counter, and found many familiar faces from my college (by many I mean a lot!). Being my third hackathon (my second Angelhack hackathon) I came all pumped about turning my idea in to reality. Farha , Akshay , jeswin, Kuriakose and I formed a team. Mr.Ratish, Regional Manager South Asia interacting with participantsRatish welcomed us to the event with a video showing what awaits the winner of the hackathon via the HACKcelerator program. Ms.Riya from IBM came and spoke about the potential of IBM Cloud to boost our ideas with their high quality products such as Watson API, Cloud Platform features and so on. She also mentioned about Call For Code, an initiative on behalf of IBM to encourage developers round the world to find solutions to natural disasters by harnessing the potential of technology. We had a good time socializing and discovering people at the venue. I got to meet a lot of people like Anusree, Renjith, Adithya, Vishal. At 12, the clock started ticking and the hackathon officially started. #### The idea To put it in plain words, we wanted to make a ChatBot that would talk to you, understand your strengths and weaknesses, find solutions to your problems and help you get better. The emotional aspect of the ChatBot would make it capable of understanding the student and provide motivational boost. To put it in perspective, a digital Assistant that is specialized for your academic well-being and remarkable emotional quotient (EQ). Apart from the emotional side, the application is capable to track your academic well-being and quantify it. I have previously tried to implement this idea at a small scale within my class. If interested you can click here to read more about it or watch a video explanation of the same.Top Articles on How Businesses are using Bots:
> 1. Five Inspirational Startups Using AI and Chatbot Technology > 2. Fifty of the best chatbot use cases > 3. How Businesses are Winning with Chatbots & Ai > 4. AI & NLP Workshop It did sound great on paper, because the next 24 hours proved it to be challenging.Day 1
The squad unitedTo start with, we decided to make a mobile app that tracks student attendance from college website, get updates from University website and converse with a ChatBot to discover student potential and find passion. So we split up; Akshay and Jeswin were working on finding a way to web-scrap the data and make it load on the app. Farha was building the app, Kuriakose was trying to set up the Google Cloud Platform and I was building the ChatBot. We initially decided to explore the potential that Watson of IBM had with their sophisticated application with features such as confidence level detection (by analyzing confidence level in tone) and much more. Since it took quite some time to set it up and get it to run, we decided to shift to Google’s Dialog flow to build the ChatBot. As time rolled by, problems slowly started coming up one by one. Jeswin was facing problems in scraping and retrieving our attendance from the college website. He went around discussing and socializing to see if someone else has a solution to his problem (investing in the knowledgeable community at the venue). Akshay had similar problems. Farha built the app real quick but had issues in integrating the cloud to its back-end. During breaks (which happened quite often) we would go around seeing the cool projects that others were working on — bus management system app, AR Sales market, Tourism made safe via validation, etc. We were intrigued by the variety of ideas that were sprouting and taking shape over there. By the evening, we felt the dead end was near because without connecting the cloud (which has the engine of the ChatBot and other details) to the app, there would not be much to showcase in the end while presenting the idea. That was when Jeswin decided to take a change in role, sat down and learnt how the back-end works using node.js . All night long he was at a corner and watching a play list on YouTube to learn the basics from scratch. Mid-night learning and socializingAfter seeing his dedication, we decided to do our share and work harder. Farha examined her code and patiently debugged it with Aswin’s help. Kuriakose on his own figured out how to connect the ChatBot to the cloud to collect and send information. I kept training my ChatBot to add more features. Here is the chart of ideas that I felt the ChatBot had the potential: Idea Chart for the ChatBotThe night was silent and many slept; the lucky ones on the bean bags while others by other means. I got to talk to Sahil from CUSAT and got to know his friends better. There was a camp fire in the night and the warmth and vibrant energy kept us up and ready. Food was regularly served and we relaxed by playing carroms and table tennis. 2AM Warmth and Joy### Day 2 We were some where between believing we could finish and feeling it would fall apart. After breakfast we had a session on how to effectively pitch our idea to the panel and points to remember while doing so. Ratish explained very well that if we followed the 30–1–30 rule (30 seconds for problem, 1 minute for solution, 30 seconds for scope) we could very well present our product under 2 minutes. It was here that we realised we had to have business plan to explain how our idea has the potential to scale up and become a successful start up. Prince Mathew, a Senior from my college who is experienced in hackathons showed us how to pitch and win the hearts of the listeners Learning how to pitch our ideaI too got a chance to present our idea, but spent too much time explaining the problem so did not get time to elaborate on the effectiveness of our product. This gave us an idea on how to better work on the presentation. In the final one hour, Jeswin and Kuriakose worked to get the back end up and running, Farha was working on better explaining the idea while Akshay was modelling the business plan. United working and brainstorming on how to fix the bugs.We decided to split up roles for the presentation. Kuriakose would be at the laptop dealing with the presentation, Jeswin would be prepared with his laptop to show a better demonstration if requested, Akshay, Farha and I took turns and presented various sections of the product. Before 12PM, we uploaded our project on the portal. You can click here to see the same and read the thoughts of the judges at the panel. The judges arrived around 2, and presentations began. We were the second team to present. Our team name and project name was ASC — Academic Support ChatBot. I spoke about the problem and the solution we proposed. Farha elaborated on the functionality and usage of the app while Akshay spoke how we planned to turn this into a sustainable business. The problem — presented by JoelThe Solution — presented by FarhaThe Business Model — Presented by AkshayThe judges asked us questions such as how we planned to generate revenue, the extend upto which the ChatBot could mimic a mentor and much more. Though we tried our best, I could sense dissatisfaction because we did not have a live demonstration and it sounded unrealistic to achieve. Neverthless, it as a good experience and we all learnt quite a bit. We also got to see the ideas of other teams —- An application that generates AR measurements of your clothes to send it to the tailor, thus saving your time.
- A sensing object to attach onto buildings that plays a pivotal role when earthquakes happen and recovery is difficult.