Blogs

Last Week Last week, I was still assembling Hermes as a practical backup for my AI development workflow. The local models were working, OpenViking was connected, and I had learned that one model slot was not enough for every request an agent system makes. I was optimistic, but the setup was not yet fully operational. This week, the memory side of that experiment began to pay off. OpenViking has not made Hermes perfect, but I spend far less time repeating the same instructions....

Last Week Last week, running out of Codex usage showed me how much I depend on AI assistance to keep development moving. I could still investigate a problem manually, but doing so took much longer than the workflow I had become used to. I still see these tools as another useful layer of abstraction. Using them well takes judgment and skills of its own. The problem I need to solve is having so much of my work depend on access to one provider....

Last Week Last week, a network problem with my self-hosted runners interrupted development. This week, I ran into another dependency that could stop me from getting work done: access to the AI tools I now use throughout my workflow. It was a short week. I did not get home until roughly halfway through it, and most of the progress was on marketing and the product launch rather than new app features....

Last Week Last week, I was working through tablet screenshots, keeping the product videos moving, and monitoring a database upgrade in integration. This week, an unexpected infrastructure problem took over: my self-hosted GitHub runners stopped getting through their jobs. Shokken is available in the US on Android and iOS. I am also preparing a reduced introductory Pro offer for the first group of users, but the subscription infrastructure for that offer is not ready yet....

Last Week Last week, I continued the app-store screenshot redesign and experimented with OpenCode as a model-agnostic coding harness. This week is a shorter update. I was not feeling especially well, but the development and marketing work continued: the third weekly product video went out, the phone screenshots reached a stopping point, and a surprising database problem turned into an upgrade rehearsal. Product Videos Keep Moving I have committed to publishing a short, customer-facing Shokken video every Thursday, and this was the third week of that schedule....

Last Week Last week, I started improving how Shokken presents itself online. The app is available to US users on Android and iOS, but simply being in the stores does not mean the listing explains the product well or gives the right customer a reason to download it. That work is still underway. The app-store screenshots in particular have taken much longer than I expected. A good listing does not just show a stack of raw screens....

Last Week Last week, I broadened the marketing surface for Shokken. Door-to-door restaurant visits had produced useful conversations, but they were too slow and draining to be my only way of finding customers. I started looking for more targeted and more scalable ways to reach restaurant operators. This week, that broad idea became two concrete areas of work: improving how Shokken appears in the app stores and publishing short product videos for the people who might actually use it....

One-Shot Coding Is the Wrong Standard I keep seeing AI coding models evaluated with a one-shot challenge: describe an entire application in one enormous prompt, wait, and see whether the result runs. That can be entertaining, but I do not think it represents a serious software-development workflow. Production software is not normally created by one person moving directly from an idea to deployment without review. There are product decisions, implementation plans, tests, hooks, code review, and deployment controls....

Last Week Last week, I paused restaurant visits long enough to reconsider how I was finding potential Shokken customers. The first dozen conversations had divided restaurants into a few broad groups. Some already had a waitlist system and no reason to replace it. Some did not have enough regular demand to justify adding another tool. Others only expected the problem to become important during a busier season. The conversations were useful, but they also showed that a list of restaurants with complaints about long waits was not a strong enough targeting strategy....

Last Week Last week made the marketing problem much clearer. After roughly ten targeted restaurant visits, I had three useful categories: restaurants with established waitlist systems that were difficult to displace, restaurants without enough current waitlist pain to justify another tool, and restaurants with visible guest waits that did not see those waits as an operational problem. The visits did not produce the first real Shokken customer, but they made the limits of Google Maps review research very obvious....