I'm 17 & I built an "Agentic AI" to kill the 15-minute Shopify listing grind
Hey hackers,
I’ve been obsessed with the operational side of Shopify lately.
One thing that consistently blows my mind: even in 2026, most merchants are still spending 10-20 minutes manually copy-pasting data to list a single product. It’s a massive growth killer.
I built Filleo to turn that into a 15-second task. It’s an agentic AI that takes messy product data (supplier sheets, PDFs, raw notes) and transforms it into a ready-to-publish listing.
It handles the titles, formatted descriptions, SEO tags, HS codes & everything else automatically.
The Tech/Strategy: I’m currently in the pre-launch phase, focusing on the "high-frequency listers" who add products one-by-one.
Try The App For Free: https://apps.shopify.com/filleo
30s Demo shows it all: https://youtube.com/shorts/K_uncUg2NVc
I’d love your feedback on:
Would You Buy It Starting At 1.99$ for basic plan?
Do You Want to try it for free to test it out? If yes then click the above URL
Does the need to clean up messy data feel like a real pain point?
Happy to answer any questions about the build & get feedback!
$0.99 doesn't make it look cheap it makes it look like you don't trust the value yet. If it saves 10 hours/week, charge like it. Even $9/mo would be a no-brainer for a Shopify merchant doing 50+ listings. The risk with $0.99 isn't low revenue it's that your early users won't take it seriously enough to give you real feedback. People who pay $0 or $1 churn silently. People who pay $9+ tell you exactly what's broken. I'd test a higher price on a smaller group before locking in the low anchor.
True and I realise that. This pricing applies only for the lowest paid basic plan. As the users feel more confident in the app and want more features, they can upgrade to pro or enterprise plans that cost 7.99$ and 24.99$ depending on their store’s size
The problem definitely resonates — that repetitive listing work adds up fast.
One thing I’ve noticed with similar tools is that accuracy and edge cases matter a lot more than speed. Even small mistakes can break trust quickly.
Curious how it performs with really messy or inconsistent supplier data — is it “publish-ready” or more like a strong draft that still needs review?
Agentic AI for hyper-specific workflows like this is exactly where the value is right now. I've been deep in the same space building AI agents for niche business automation and the pattern I keep seeing is: the more narrow the use case, the better the product-market fit. Shopify listing automation is a perfect example — it's a repeatable high-friction task that merchants deal with daily.
One thing I'd flag from my own testing: make sure the data extraction handles edge cases well. Product descriptions with weird formatting, variant combinations, and images from different sources are where most automation tools fall apart in practice. Would be curious how you're handling those.
Agentic AI for hyper-specific workflows like this is exactly where the value is right now. I've been deep in the same space building AI agents for niche business automation and the pattern I keep seeing is: the more narrow the use case, the better the product-market fit. Shopify listing automation is a perfect example — it's a repeatable high-friction task that merchants deal with daily.
One thing I'd flag from my own testing: make sure the data extraction handles edge cases well. Product descriptions with weird formatting, variant combinations, and images from different sources are where most automation tools fall apart in practice. Would be curious how you're handling those.
круто
That youtube link is broken btw.
GOOD idea bro , keep going
Really appreciate that!
Interesting approach. I’m currently building something for workshops and I noticed a similar issue around visibility and operational tracking. I think the hardest part is not building the tool, but getting real adoption from offline businesses.
Very true.
Love this the 10-20 min listing grind is painfully real Would definitely try the $0.99 early access
One thing in your demo maybe show the messy before data more clearly Most people will just need to see oh thats MY problem for 5 seconds to get hooked
The Shopify merchant pain point is massive Rooting for you
This is a strong direction, but the real value isn’t just saving 15 minutes per listing. The deeper problem is the mess behind product data itself—supplier sheets are inconsistent, formatting is manual, and merchants lose energy trying to maintain quality while scaling. What you’re really building is closer to a product data standardization layer than just a listing tool.
Where this gets interesting is if it goes beyond generation and starts improving outcomes. Imagine it learning what kinds of titles convert better in a category, or adapting descriptions based on performance, or even cleaning and structuring supplier data into something truly publish-ready without human correction. That’s where it shifts from “AI that writes listings” to a system that understands e-commerce inputs and outputs.
The pricing experiment is bold, but be careful with signal quality. Ultra-low pricing often attracts curious users rather than serious high-frequency sellers. You might get better insight from usage-based pricing or a free trial that only converts once real value is felt through repeated use.
Right now the framing is “faster listings,” but a stronger positioning is “removing the bottleneck in scaling product catalogs.” That subtle shift moves it from a productivity tool into infrastructure for e-commerce operations.
The key question going forward is whether you’re optimizing for speed of listing, or for the performance and scalability of the products being listed. That choice will define how far this can go.
The "AI Auto-fill" pattern resonates strongly — I've shipped a similar internal tool for a different domain (video reference cataloging), and the same dynamic shows up: the human + AI loop is dramatically faster than either alone, but only when the AI prompt is dynamic to the current taxonomy state, not a frozen static prompt.
One thing that took me embarrassingly long to figure out: parse the AI's JSON output INCLUDING markdown code blocks. Models keep wrapping JSON in fenced blocks even when you tell them not to, and brittle parsers break. Had to add a tolerant extractor that strips fences before JSON.parse — saved us a bunch of "regenerate, please" cycles.
Curious how you handle the moderation step on the Shopify side. We have admins reviewing AI-filled records before they go live, and the failure mode I see most: AI confidently fills a field with a wrong-but-plausible value (e.g., wrong product category) that admin clicks through without re-checking because everything else is right. Have you seen that on the listing-fill side? Any UI pattern you found that flags "AI was unsure here"?
Either way — built like an actual product, not a demo. 17 and shipping like that is impressive.
curious what happens after PH day 1. most indie tools get the spike then silence - zero existing users means no baseline for who sticks and who just clicked out of curiosity.
This hits.
I’m building small utility apps and I see the same pattern:
people don’t quit because the task is hard,
they quit because it’s repetitive.
15 min tasks repeated 100 times = no growth.
Your angle makes sense.
Curious:
Are your early users actually using it daily,
or just trying it once?
Retention here will decide everything.
$0.99 specifically signals "I'm not sure this is worth more", even if you don't mean it that way.
The psychology: buyers don't anchor on what they paid. They anchor on what you charged them to decide.
$0.99 says the decision was trivial. $9 says the problem is real.
Same friction to sign up. Very different signal about what you believe the tool is worth.
Very true. 0.99$ is to get the user started. The user can later switch to more advanced plans like pro or enterprise which cost 7$ and 24$, offering more features as the user feels more confident in investing in the app
That's the question that matters. Funnel entry only works if the upgrade is inevitable, not optional.
If $0.99 buyers can get enough value to stay at $0.99 forever, you've built a $0.99 product. The cap has to feel real before they hit it.
$0.99 might be too low for the buyer profile you actually want. Someone running a serious Shopify shop with 100+ monthly listings would pay $30-50/month for this without thinking. The "$0.99 first 100" framing also makes it harder to raise prices later, those buyers anchor at the cheap rate.
If the goal is launch traction, founding members at 50% off forever is a stronger mechanic than $0.99. Same urgency, doesn't kill your future pricing.
Also pre-launch and figuring out similar tradeoffs. Curious what made you land on $0.99 specifically vs say $5 or $10.
Really appreciate the feedback. I’m trying to test user response, and will raise the prices once I get initial traction. Moreover, this is the price for basic plan. Pro and enterprise are 7$ and 24$ forgetting more features and credits
That makes sense, the $0.99 is the funnel entry not the actual price. Worth thinking about whether the $0.99 buyer ever upgrades though. In my experience tiered pricing only works if the entry tier is intentionally limited enough that serious users hit a wall and need to upgrade. If the $0.99 plan is "good enough" for your high-frequency listers, they'll stay there forever.
What's the gap between basic and pro that pushes the upgrade?
a 0.99$ subscription gets you much fewer credits (~100) product generations per month, while a 7$ one will give you much more than that. Obviously if it is a mid-tier store, just generating 100 products in a month might not be enough and they'll want to upgrade. I do realise that while there might not be any additional features unlocking as the user upgrades to next tier, I wanted to keep my app simple and to the point
Got it, credit limits are doing the upgrade gating instead of features. That's cleaner than the SaaS pattern of locking features behind tiers, you avoid the "I'd pay more if it had X" objection. Good call on keeping it simple.
Nice, how did you get started?
I created my own shopify store at 16, but soon realised that it was really tough to upload each product listing by hand onto the store as a new reseller. I tried chatGpt but it was inconsistent and not personalised at all. Moreover, it sometimes made mistakes in processing messy data as well. So I decided to make a Shopify app to solve this problem, because I realised that many more like me might be facing the same problem
Nice problem to tackle — this is real pain.
$0.99 will get signups, but also low-intent users. You might learn more from fewer people paying $5–10 and actually using it daily.
Also curious: how consistent is the output across messy inputs? That’s probably the make-or-break here.
Yeah it is an early pricing to test user response. Also, this is the pricing only for basic plan, the pro and enterprise are 7$ and 24$
i would buy it bro
Really appreciate that! Feel free to join the waitlist above if you’re interested to get early access
Wow loved the idea. Great to see young entrepreneurs like you change the world like that! Signed up to support ya