AI Dev 2026: Reflections From a Conference I Wasn't Ready For

I'm Back!

I've been back home for a couple of days and have taken a little time off to reflect on how Deep Learning's AI Dev 2026 conference went. First and foremost, I had a great time and I'm very happy I took the time to go. I know in my update post I talked about how nervous I was, and that only got worse when I arrived in San Francisco. This anxiety was only worsened when, to my excitement and chagrin, I was emailed the night before about winning an award at the conference. I won an award for most engaged learner! Turns out I had more certificates than any other attendee, and they wanted to announce that during the opening keynotes. Little did I know, but apparently 23 certificates and 153 hours in Deep Learning was quite a lot. Obviously, I was excited and honored to be recognized, but still nervous about being named in front of so many people, even if it is a very good opportunity. To my abject horror, I wasn't just named — I ended up coming on stage, meeting Andrew Ng, and accepting my award. I have never stood in front of so many people. I survived, but wasn't prepared for being recognized for the rest of the conference.

My Brief Flash of Fame

The one thing I was not prepared for at all was the number of people that wanted to talk to me. After all of the opening keynotes, I had quite a number of people coming up to congratulate me about my award. People were shocked and excited by the sheer number of certificates I'd received. On day one, I was definitely too shy about the attention. At the end of the day I told myself that I should embrace the compliments and attention and use it. I definitely wasted some opportunities to talk to people on day one. Accepting the attention on day two changed everything. I had so many people coming up to talk to me once they realized they could approach me that I spent almost an hour and a half with a constant stream of people ready to talk and ask me questions. I met some very cool, interesting people and learned that business cards aren't much of a thing anymore. Everyone asked me for my LinkedIn, which I have but don't use often, while I had had business cards made to hand out.

Feeling Validated

Uncertainty and lack of confidence were apparent when I arrived in San Francisco. By the end of the conference, I felt extremely validated in all of the hard work I'd been putting in. When people would come up to talk to me, I was shocked by the number who wanted my advice or opinion. In my head, I kept asking why they were asking me — I had no idea what I was doing. The more I talked to people, the more concrete my answers became, and the more I even started to understand what I'd been doing. It felt fun and engaging to talk to people about AI and the courses I'd been taking. I had a lot of people asking about courses they should take, which ones I liked, and what they could do to improve. The shock on people's faces when I told them I was a server in a restaurant was fun — also encouraging because of the number of people who thought that was really cool and inspiring. I was even told that I explained some concepts we were learning about in one of the talks better than the lecturer. A couple of the people I talked to the longest even heard my long-term company goal and thought it was exciting and unique. That helped a lot in me feeling validated for all I've been learning. While the fame was extremely unexpected, it really helped me to enjoy the conference and, more importantly, feel validated in my direction.

A Happy Side Quest

The theme of the conference was memory and agents. I attended many classes on both and came to an interesting conclusion: those are not necessarily things I want to focus on. I really enjoyed a lot of classes, but a lot were either not in-depth or technical enough, or they were a little too specific for anything I wanted to do. I know I've been wondering what to do with my company because I'm a long, long way off from my final goal. The idea of building things for small businesses sounds like a good idea, but one I'm not very confident in. The conference did open my mind to two things. I've always been a good teacher — I've trained or taught at every job I've been in and every major hobby I've done. Secondly, the one thing I've been doing for the last two years has been learning, and more importantly, how I've been learning with the help of LLMs. So, while building agents and automations feels overwhelming to me, I can definitely feel inspired to help teach people about AI. I think starting small by teaching people how to use these tools to help them learn or improve their life at work sounds a lot more approachable. While doing that, I can start to learn how to build small things for people that I can help them with once I've taught them the basics. This isn't a switch in directions — it's more of a starting step for someone who's unsure about the next couple of steps.