How I Won A Gold Medal at NaijaHacks '19
Happy New Year, folks! This is my first post this year.
It's been over a month since we won Gold at the biggest hackathon in Nigeria, NaijaHacks, 2019 edition, and I apologize for writing late. The usual me would have forgotten about writing about my experience, but I have developed this mindset that a project is not truly finished until I'd written a documentation on it.
I wrote about the very draining, a week long, online hackathon in an earlier post. This article will only been concerned about the Expo day.
An expo is a large exhibition event where goods and products (in our case, projects) are displayed.
THE PROBLEMS WE AIM TO SOLVE
- Plant Disease Detection
- Plant Pest Detection
- Animal Disease Detection
- Soil Fertility Level Detection and
- Crop Recommendations.
OUR PROJECT IDEA
My team, Team Perspective, worked on a project we named AgroAI. AgroAI is an app built for agriculturists and youths wanting to delve into agriculture. It is equipped with four basic functionalities;
- Plant Disease Detection
- Plant Pest Detection
- Soil Fertility Diagnosis and recommendation of viable crops.
- Livestock Maintenance.
We also planned on building a forum, a community, where farmers from different parts of the globe could network, connect and discuss best agricultural practices.
This idea later evolved. We finally decided to integrate it with existing popular platforms; Nairaland - Agricuture Section as a local community and Community.agriculture.com as a global community, to save us the stress of having to build a forum from scratch.
HOW OUR PROJECT IS DIFFERENT FROM SIMILAR/EXISTING SOLUTIONS
Of course, there are apps that do the same thing our app is supposed to do. But those apps have issues ranging from internet connectivity, network latency to complex GUIs. And this is a potential issue for farmers in Nigeria as more than 75% of farms in Nigeria are located in rural areas where internet connectivity is usually poor.
That was what made our app unique. A farmer would be able to carry out a disease diagnosis on his plants in his rurally-located farm without having to worry about internet connectivity.
Since our app is designed to work offline, there won't an issue of network latency which usually occur when numerous requests are being made to a server simultaneously. (Network latency is the extremely slow internet connectivity that you experience when bidding for hostel allocations in your school).
Read more about AgroAI here.
WORKING ON OUR PITCH AND SLIDES
We did this the night before the expo day, and it took up the whole night. I remember not sleeping until 5am, and I'd to wake up by 7am for the expo day preparations.
THEN IT STARTED…
The event started an hour and half later than the stipulated starting time. The founders of NaijaHacks, Uche and Blessing, gave the welcome address and the mission and vision of the hackathon. There was a keynote speech from Chude Jideonwo, the founder of Joy Inc.
We had our breakfast not so long after, and the pitching began!
Tables and corners were arranged with each team at her table while the judges (already assigned teams' numbers randomly) moved around and listened to each team's pitching.
Our first pitch was a horror show. Our nervousness was so obvious and our judge ripped us to pieces. Folahan, my teammate, is an excellent guy with superb coding abilities. But communicating and handling presentations with non-technical people is not his strong point. When the first judge left, we reworked our pitching style. Folahan took up explaining the technical side of our project and I focused on selling it and explaining why our project is the best thing that has happened in Africa in December.
Of course, our four other pitching sessions were a success.
Here is a picture of a team(not ours) pitching to a judge.
I couldn't believe it when AgroAI was announced among the best 12 projects. With how horrible our first pitch session was, I never thought we could rank among the 12 projects at NaijaHacks '19. The only reasonable explanation is that we must have ranked crazily high on other judges' sheets.
I mean, I walked around and saw lots of wonderful projects, yet we still manage to scrape through as the one of the top 12 projects.
Below is a picture of the top 12 teams. My teammate and I are at the extreme lefthand side.
There was a final pitch session on the stage to pick the best pitch among the top 12 projects. After that, the winners of the different award categories were announced.
I got bombarded by questions after our pitching. Some hours after the event ended, I saw some complaints on the Slack channel for NaijaHacks '19 and I choose to address every one of those complaints that apply to AgroAI.
- Are you truly a beginner?
This was the first question I was asked. Yes, I am a beginner. I started programming around October, 2018 with lots of blocks and stops in between. But Folahan is not a beginner.
NaijaHacks' rules state that a team can register as a beginner team if 50% of the team members are beginner. So I think, we qualified as a beginner team.
- How could you finish a project of that magnitude within a week?
During our pitch on stage, I mentioned that none of our team mates had any prior knowledge on mobile development yet we manage to build a working prototype of our project on Android and iOS platforms.
Yes, we could. With purposeful steps and a keen sense of direction, it is almost easy to accomplish a project within a week. Also note that all that is required from us is a MVP (Minimum Viable Product).
The first day of the competition was for team formation, Bizskills training and brainstorming.
The second day, we had more brainstorming. Then Folahan was working on our UI mockups while I was sourcing for needed datasets to implement our ML models.
The third to sixth day, Folahan was brushing up his Java knowledge and working on the Android and iOS apps while I was battling with Continuous Assessment Tests which I couldn't prepare well for. Surprisingly, the Android version gave him much trouble than the iOS.
By the seventh day, we were building the neural network for the crop disease detection functionality.
Before the midnight of the final day, we were battling to fit our model in the apps and submit few minutes to the deadline.
We really needed a mobile/web developer on our team so we could build a web app and the forum functionality we wanted to include but we couldn't find a suitable one.
An organization that we applied for, to release data for us to build the soil fertility detection functionality. didn't get back to us throughout the hackathon's building phase.
So yes! We built AgroAI within a week with lots of difficulties and determination.
- How could you be one of the top 12 teams when your project is not in any of the stated categories?
The stated categories for the competition were Education, Housing and Accommodation, Job Creation and Citizen Safety.
AgroAI is a project that focused on agriculture and we had people complaining about this.
No, you were wrong. Our app belonged to not just one, but two of the stated categories.
Folahan mentioned while we are pitching that there's a time an agricultural department in our school approached his department to help build an app that does the same functions that AgroAI does (Folahan was supposed to be part of the team that was to work on the project). But due to internal strife and politics, an agreement couldn't be made.
Thus, AgroAI qualifies as an Educational project, as it would be used for research in (at least) all agricultural colleges.
AgroAI also qualifies as a Job Creation project, because it would make lives easier for farmers while also encouraging youths to venture into agriculture as a career.
Now some might argue that encouraging youths to venture into agriculture does not necessarily create new job opportunities. I might also argue that matching job seekers with companies with vacant spots (which was what most of the Job-creation projects were all about) does not necessarily create new jobs.
Let's be clear, though. AgroAI does not excessively fit into any of the four categories. And this hurt our chance of winning any of the categories, which I was a bit disappointed about. I really fancied winning a laptop. I need one, badly.
WHAT I LEARNT
I feel people who would participate in future hackathons might need this;
- Your soft skills are just as important as your technical skills.
I saw lots of fantastic projects which I thought were going to be selected for the top 12. You can't imagine how surprised I was when they didn't make the cut.
The judges, especially when they are business-oriented folks, don't care about the programming languages you implemented your idea with. They don't care about the tools and technologies you used.
All they care about is the business proposition that your idea offers, and how practicable it is.
While on the bus back to our hotel, there was this guy who wasn't impressed with the judge's choice for the Citizen Safety category's winner. He said the technology his team made use of was superior to the winner's, as it was initially implemented by a top-notch Russian security company, blah. Inasmuch as your technical skills is important, don't waste time explaining the technical process behind your project unless the judges specifically ask. I think if judges can relate very easily with your ideas and solutions, it makes things easier for you. And you score points more easily.
- Why You?
Knowledge, Skill, Passion and Experience. Find the intersection where you stand.
If you can relate your problem solution to your background/personal experience, it makes people believe that you are the best fit for your project. While pitching our ideas to the judges, our being students from an agricultural college (albeit we are non-agriculturist students. I was a Computer Science Sophomore while Folahan was a Mechatronic Engineering Finalist) aided our legitimacy to work on a project that solve agricultural problems.
You need to project The Messiah image to the judges. Showing how your background, experience, passion and skills intersect with your idea makes the judges believe that you are the best man to solve the problem, thus, earning you credibility and more points.
- The most eloquent teammate should handle the bulk of the pitching.
I think this is self-explanatory.
You don't need to be a bad ass programmer before you can create problems and come up with solutions. I've seen people who limit themselves due to their perceived lack of technical skills. I also met a team with no prior programming skills picked up 2 awards (Top 12 projects and Best Pitch Award) at NaijaHacks '19. You can always pick up the required skills on the road.
Remember the popular saying, "Programming is easy. Problem solving is the hard part."
Happy Hacking, folks.