Jarvis, honestly
Personal · 2026-08-12
Building · 2 min read
Jarvis, honestly
So I want to talk about Jarvis. Not the polished version, not the version I'd put on a landing page with a nice gradient and three fake testimonials, just where it actually is right now.
Building in public only really works if you're honest about the bits that don't work yet, right. Anyone can post a screenshot of a win. That's easy. The harder thing, and I think the more useful thing for anyone reading this who's building their own product, is talking about where it's rough.
What's actually working
Jarvis is good at the stuff I built it for first, validating whether an idea has legs before you spend three months and your savings on it. Feed it a concept, it pulls together market signals, competitor gaps, pricing sanity checks. The bit that surprised me, honestly, was how well it handles the "is anyone actually searching for this" question. That's the question most founders skip because it's boring, and it's also the question that kills most ideas before they get anywhere near $1M ARR.
Users have told me the validation reports save them a genuine chunk of research time. Not hours of magic, just... less faffing around with fifteen browser tabs open trying to piece together whether a market exists.
Where it falls short
Now the honest bit.
Jarvis still struggles with nuance in niche markets. If your idea is a fairly standard SaaS play, it does well. If you're doing something genuinely weird, some hyper specific tool for, I don't know, artisan cheese wholesalers, it starts guessing more than reasoning. I'd rather tell you that than let you find out the hard way.
There's also a lag issue with real time data. It's not instant. Sometimes the signals it pulls are a few days stale, which matters less for long term validation and matters a lot if you're trying to catch a trend as it's happening. I'm working on it. Anyway.
And the interface, honestly, needs work. It's functional. It's not pretty. I built the backend first because that's the bit that actually matters for accuracy, and the frontend is very much a "does the job" situation right now rather than a "delight" situation.
Why I'm telling you this
Because the indie hacker crowd, the bootstrapped builders, the people scaling something from zero, they don't want marketing copy. They want to know if the tool actually does what it says, and where the edges are. I'd rather lose a sale from someone who wanted magic than gain one from someone who feels misled a month later.
There's also just something freeing about saying it plainly. I mean, I built this thing, I'm proud of it, but pretending it's finished would be a lie, and lying to your own audience is a slow way to lose them.
What's next
Roadmap wise, the priority is tightening up the real time data lag and improving how it handles niche or unusual markets, since that's where most of the current friction lives. Interface improvements are coming too, though slower than I'd like, because accuracy always wins the priority fight in my head.
I'm not going to pretend I know exactly when any of this lands. Software timelines are famously optimistic and famously wrong. But the direction is clear: make the validation sharper, make the data fresher, make the thing look less like a spreadsheet had a baby with a form.
Anyway, that's Jarvis, warts and all. If you're building something similar, or even something completely different, I think the lesson holds either way: tell people what your thing actually does, not what you wish it did. It's a much better foundation to build trust on, and honestly, it's a lot less exhausting than keeping up appearances.
That's it for today.