sankalp phadnis

lemma's "wing it" mode lets you go from zero to a first draft in minutes. you answer 6–9 quick mcq questions about your feature, and the agent takes over — reading your context docs, researching the ecosystem, and writing a full prd.

the research step is genuinely useful, but it takes a bit. rather than stare at a spinner, i added a chrome dino game to the loading screen.

small detail, but it's the kind of thing that makes a product feel like someone actually uses it.

every founder tracks monetary runway. months of cash left, burn rate, time to next raise. but there's a second runway that nobody talks about: energy runway. how many months of focused, creative work you have left in you before the quality of your decisions starts to degrade.

i learned this the hard way. after building and pausing zarie, i jumped straight into lemma — an ai-native prd editor. the idea was strong, i was shipping fast, and my bank account was fine. but a few weeks in, i noticed the signs: shorter attention spans, more time questioning the path than enjoying the work, less excitement about ideas that should have excited me. i had been careful about money, but i'd completely ignored energy.

in retrospect, the drain had been building for a while. i'd mostly built in isolation — didn't build in public, didn't lean on a community. i kept telling myself that if i just thought harder and shipped more, the breakthrough would come. instead, my world shrank to a loop of building, doubting, and pushing harder without changing how i was working.

monetary runway is easy to measure. energy runway isn't, which is exactly why it's dangerous. by the time you notice it's low, you're already making worse decisions — picking the wrong problems, over-engineering, avoiding the hard conversations with users. the startup doesn't die from lack of cash. it dies from a founder who's technically still working but no longer really building.

if i could go back, i'd track energy the way i tracked money: deliberately. that means building with other people, not just for them. sharing work early, even when it's rough. taking breaks that actually reset you, not just ones that pause the anxiety. and being honest about whether you're still pulling toward the problem or just pushing through out of obligation.

your startup needs you sharp, not just present. protect both runways.

we started zarie from a deeply personal need. we've always struggled with executive function: remembering complex supplement schedules, staying in touch with family and friends, and generally keeping on top of the small but important things. we wanted an assistant that lived where we already spent our time and quietly made sure nothing slipped.

we built the first version on telegram. under the hood, it used a multi‑agent architecture: a main agent (zarie) that could invoke specialized worker agents for specific jobs. in practice, that meant we could have a health tracker agent, a social accountability agent, and others, all coordinated through a single, continuous chat. the experience was intentionally single‑threaded so it felt like talking to one real assistant with memory, not a collection of disconnected bots.

on top of that, we built a reminder and automation system for time‑ and search‑based workflows. you could set up things like, "remind me before every india match" or "ping me before every f1 race," and zarie would handle the recurring logic behind the scenes.

we scaled zarie to around 100 users (mostly friends and family) and watched their behavior closely, along with our own. that's when a hard truth became clear: while people liked the idea, it wasn't a hair‑on‑fire problem. the number of genuinely valuable, recurring automations people set up was small, and most users didn't naturally think in terms of creating new workflows. even when we brought zarie into slack — where people already spend their workday — the pattern held. usage was sporadic, not habitual.

at that point, we had to be honest with ourselves. we had built something we personally enjoyed, with an interesting architecture and real users — but not a product with urgent, pull‑based demand.

rather than continue to push a nice‑to‑have assistant uphill, we decided to pause zarie.