Startups

Some did good, many didn't. Here are the products I've built, and for the ones that didn't make it, why they failed.

● Succeeded / Active  ·  ● Shut down

Carve

Carve

Active

Client-ready architectural renders in 60 seconds. Carve turns your sketches, CAD exports, and property photos into polished, client-ready images. No render studio, no week-long turnaround.

Locus

Locus

Active

Building the present and future of the business-travel economy. Locus partners with events to provide travel services to attendees and teams.

Viraloop

Viraloop

Active

Create your own AI influencers that generate and auto-post viral content to TikTok, Instagram, and YouTube automatically. Set it up once, go viral on loop.

ScreenTag

ScreenTag

Shut down

A tiny macOS menu-bar app that tags each Mission Control Space with an emoji or icon. Put a 💻 or 🎨 on a Space so you always know which workspace you've swiped into.

Never marketed.

SitFit

SitFit

Shut down

A native macOS app that helps remote workers stay healthy during long sessions through intelligent break reminders, posture monitoring, and eye-strain prevention.

Never marketed. It stayed a hackathon project.

Festify

Festify

Active

The tech partner for distinguished events. We build and manage the TOKEN2049 Week websites (Dubai & Singapore) and the SuperAI Week site, plus projects like TOKEN2049 Dubai Badges, the HODL Challenge tweet bot, and SuperAI Badges.

Faceless

Faceless

Succeeded

Faceless.so turns text (prompts, blogs, even Reddit posts) into faceless videos for TikTok and YouTube, complete with captions, voiceovers, niche customization, and automated daily scheduling and publishing. I handled everything from DevOps to frontend, took it from 0 to $100k+ in revenue, and acquired slickwid.com (rebranded to "Story" by slick.is).

Slick

Slick

Active

Turn YouTube videos into shorts and reels. Auto-add trendy captions, cut silences and filler words, snap in b-roll, sound effects, magic zooms, and more.

Idea2Business

Idea2Business

Shut down

A tool that turns a one-line idea into a growth plan in under a minute, with a custom pitch-deck template to kickstart fundraising.

No real market demand. The target users could already do most of this for free through ChatGPT.

Get Into Web3

Get Into Web3

Shut down

An education initiative to help people break into Web3, blockchain, and crypto through blogs, hands-on training, mentorship, and a community.

Not enough motivation or expertise in the team, and no clear vision for how to move it forward.

Musixverse

Musixverse

Shut down

A music NFT platform built on the ERC-2535 Diamond and ERC-1155 upgradable standards, with a sharing mechanism that paid the sharer a cut of each NFT trade. I led a team of 7 core members, 3 interns, and 5 advisors. I owned the Node.js/Parse backend, MongoDB optimization, AWS cloud ops and CI/CD, and integrations (Twitter, Persona, SendGrid, Stream Chat).

It was the wrong kind of startup for us. It needed funding, network, or deep experience, and we had none of the three. Most of the team couldn't go all in around day jobs and college without enough compensation, early traction never picked up, and Web3 and NFT friction forced users to learn too much before they could even use the product.

MrktDB

MrktDB

Shut down

A platform that aggregates SEC 13F filings to surface hedge-fund holdings and trends through interactive visualizations. Built with Django REST APIs, React + Chart.js, Python scrapers (BeautifulSoup/Selenium), and scheduled cron jobs, deployed across AWS (EC2, Elastic Beanstalk, ECS, S3, Lambda). Also shipped the open-source failstodeliver Python package.

The team didn't have the experience at the time to build something this complex. There was a lack of communication, and people kept skipping meetings.