Hey everyone! I'm currently a second-year software engineering student, and I recently went through an intense internship interview. The interviewer challenged me on various topics, including RAG-based chatbots and unit testing. While I did my best to answer, he pointed out that I seemed to rush into building without fully understanding the project requirements or having a solid research methodology. This really stuck with me, so I wanted to ask how you all approach research for a project before diving into the development phase. What resources do you utilize? How do you incorporate AI into your research, and how do you ensure you're learning what you need for your projects?
5 Answers
Honestly, who cares about massive research for personal projects? Get in there and start building! If you find something doesn’t work, that’s when you learn. I love making stuff and not every project lasts long enough for early decisions to really matter. Just have fun with your projects, that’s what counts!
If it helps, I've been in this field for over 30 years and I often start coding without fully grasping everything either. It hasn’t hindered me much! Sometimes just diving in and figuring it out as you go can work too, especially with personal projects!
It sounds like the interviewer was really pushing you to understand the client's needs. When I start a project, I always look for similar existing solutions. If my idea is unique, I'm okay with spreading some creative wings and figuring out what new needs to be developed. And if I'm recreating something that's already out there, I don't feel bad about it; we all go through that phase!
Research really depends on the project type. For me, the first step is discussing with product management to clarify the requirements. Then I think about which teams need to be involved and set up meetings to align on timelines. I also spend time prototyping and testing things out, sometimes even reading through academic papers if I’m venturing into unknown territory. It’s all about gathering what I need before jumping in!
Exactly, the documentation you generate can really help keep everyone on the same page.
Managing a project isn't just about the code. In my last job as a PMO head, I learned that once you're assigned a project, you need to think comprehensively. Instead of just diving into coding, first assess the entire scope and then break the project down into manageable parts. Once you have that plan, get the resources lined up and only then start building!
Sounds like a solid plan! It's important to cover your bases early, especially to avoid those last-minute surprises.