One of the things I learned the hard way in month 1 (July to August) was that creating content, designing, and coding at the same time means doing them all horribly.
Since the announcement of Project #2 - Privately, I haven’t been writing as much and I’ve been fully dedicated to coding. The good news is that I spent about 2-3 weeks neck-deep in Claude Code, CODEX (OpenAI), and GitHub Copilot, and I have finished the basic front-end. local-first back-end, local AI implementations, and local encryption. I completed these tasks within a week. Then the hardest part came.
Integration
I never thought that piecing together the front-end and back-end would be so difficult, but it’s a lot more complex than I expected. The database got corrupted due to CRUD (Create, Read, Update, Delete) functions in the front-end. Some were due to overlapping messaging. Others were caused by running simultaneously, or running out of order, then the entire database needs to be deleted. Of course, this can never happen in a real-world environment where people are storing their precious journals, so I had to find a permanent solution for this.
This has been ongoing for 2 weeks.
I’m finding the solution now to ship this v0.1 that I wanted to be fully able to test with friends on 09.10.25. The delay is directly attributed to the lack of knowledge and tools. Due to the complexity of the problem, my toolkit and learning have grown significantly. I had no time to focus on anything else. I will share all learnings with you all in the coming posts.
Managing Sprints
It’s been a strong 5 weeks of engineering learning. In addition to the basics that I’ve learned in 4-6 weeks in July - August, I can truly say I’m now coding coding. Not just Vibe-coding. One key learning is that, as I go deeper in software engineering, the more focus it requires. More and more complicated topics emerge, and more trade-offs and architectural decisions that I have to make—AI can’t make these decisions for us.
This means I need to manage my time and focus better, in sprints.
I believe every founder should implement these sprints for themselves.
1 Week Sprint - Choose 1 of 3 - Code / Design / Share
In any given week, I have decided to focus on one of the 3 areas: Code, Design, Share. Unfortunately running a business means we can’t leave everything else on pause, so we have to choose to maintain our progress on other fronts. The reality of a business push us to always share. Distribution is becoming more important than our ability to build & design these days. Therefore, for any sprint it looks like:
Code + Share
or
Design + Share
Admittedly, I haven’t been doing this. I have been just straight coding nonstop for 5 weeks. Not posting much on substack or any social media. For a single person to context switch between 3 different areas in a single day is impossible. There were many days where I got distracted, then I got overwhelmed, and I went back to coding. This stops this week.
I’m finishing my Coding Sprint, looking to have a live v0.1 for myself and a few friends to test before sharing.
The hardest part is creating content while managing the sprint. I have some thoughts on how to improve this.
I have a few days to share my findings from my engineering sprints before the design sprint, and I will do my best to not over-edit. I’m trying really hard to not kill my voice in the editing process, which I can see that I have done in the past.
The upcoming posts:
Privately Alpha v0.1 Launch & Future Plans
Modern Coding w/ AI (Guide)
The Next Level of AI Coding (In-Progress)
100 days of Coding Lessons & Mistakes
Next week, I’m focusing on design. UI / UX. Animations. And also focusing on ramping up my social media / writing efforts. If you don’t hear from me, please yell at me to keep me accountable.
Traveling hasn’t helped in terms of keeping me disciplined, but new challenge accepted.
Here’s a picture of me coding the last 2 weeks.
- H