Work on a Terrible Idea

But I still did it

"I think he will HATE this idea," I said.

My boss was in France, so he wasn't part of this conversation. But I knew he only appreciated my contributions when they led to immediate gain.

"He will probably not like it. But you need to do it, and in the end, show him a bigger universe; he will understand," my colleague K encouraged me.

My terrible idea

Our product is using AI to create a lesson based on user input. This idea is to improve the next lesson based on the current one. As of now, no one uses it because the AI-generated next lesson is always boring and repetitive to the original lesson.

So we decided to improve upon it, with the assumption that the next lesson should be more challenging than the first one.

Hence, every day in the standup next week, I mentioned my work to my team, but I don't think anyone except K understood it. (I don't think people really listen to each other during the standup anyway.)

Weird humans, whether boys, girls, or them, this is a huge risk for me. My boss already thinks I'm not "working hard enough." I also didn't know if I'd get a return offer. But now, I'm trying something that might not work. And even if it does work, he might still not like it. But I really want to give it a go, even if it means I might lose my offer because of it.

A swing and a miss

I worked hard on this new idea for the whole week. The things I found out were really interesting but hard to explain to others. The way AI learns is very different from the way we learn, and that makes it tricky.

Moreover, at the first standup when I was excited to share my work, I got so many craps. 🚽

"This looks like a bug!" one voice echoed through the room, shattering the delicate veil of hope that had encased me. "You shouldn’t do it on production," another chimed in, the words landing like stones upon my chest.

I realized too late that I hadn’t focused enough on making the user interface (UI) look good for our product. Damn, such a horrible first impression. I had to spend the rest of the day undoing everything I had done. Painfully.

Let me try again

But I haven’t give up, that night, I spent the night crafting the storyline before our all-hands tomorrow. Preparing for the critiques that might come after me. I don’t know if they will find my presentation a waste of time now. But I need to try it anyway. At most, they couldn’t appreciate it and threw it away, telling me “I’m not doing good work” again. But I had to stand before them, for my passion for AI and the boundless realms of learning it could unlock. In the morning, I presented with many examples and analogies to help them digest. However, no sooner had I presented the first slide, than my founder interjected, "But your approach seems to only dive deeper; what about providing breadth?"

I was prepared for this question, and with a reassuring tone, I urged “Wait for a little bit and I will show you.” On the next page, I immediately showed how AI can provide breadth even with my dive deeper methodology.

Drawing towards the conclusion, I held their gaze and said, “Look. I’m not only creating a better next lesson.” I paused a little. I paused, letting the words hover before I continued, “We’ve been an app focused solely on unit courses until now. But now, AI is able to create high-quality new lessons that are connected with the previous.

Picture the realm of knowledge as a universe. Each tap on a star now has the potential to light up other connected stars. AI keeps throwing surprises our way, enabling users to create through their own, unique universe.”

I paused again, letting the vision seep into their minds before posing the question, “How do we want to design this interconnected experience for users?”

As the last word escaped my lips, I surveyed the room, feeling unsure if they would appreciate it.

Then, like a gentle rain, the applause began. The faces that once bore skepticism now sparkled with awe.

Reflection

I don’t know if they truly understand my implication of non linearlity of AI through my analogy, but I know everyone is thinking when I talk. And they recognized my contribution. Immediately in the afternoon, my founder told me we could start our offer process.

I wasn’t entirely sure if they fully grasped AI’s fascinating non-linearity through my analogy, but I could see the wheels turning in their heads as I spoke. They were pondering, and analyzing, and that in itself was a win. And they did recognize the effort anyway! By the time afternoon rolled around, my founder told me we could start our offer negotiation process.

All of these need to thank to my colleague K. He encouraged me not to appease my founders, but really nudged me towards great work regardless of potential failures or critiques from others.

On the surface, improving the next lessons seems to be a terrible and low-value idea. But in the end, it unfolds so many possibilities. It was like peeling layers off an onion, each layer revealing a deeper, more profound understanding of learning and the potential our product held. And this is only the beginning.

What I learned

• As Paul Graham and Sam Altman said, teams eliminate both the worst and the best ideas. To allow room for great ideas, I need people like K to provide a safe space for me to express my ideas freely without judgment.

• Even if I am a teammate and not a founder, I need to sell my ideas and sell them really well through preparation.

• Often, people struggle to envision innovative ideas. It's only when you've completed your creation and shown it to them that they learn to appreciate it.

• Though an idea may appear terrible, its output can potentially generate the next good idea. Take action and risk to experiment!

Note: Thanks to Shani & Svit for the edition