Pieter Levels, known as @levelsio, builds and runs internet businesses including Nomads.com, Remote OK, Photo AI and Interior AI. His project archive and blog document a long series of bootstrapped experiments, including the many that did not become lasting businesses.[1][2]
Start with List of all my projects ever →
Education
Links
Website
X Founder · Photo AI
Founder · Interior AI
Founder · Nomads.com
Founder · Remote OK
Background
Levels studied Business Administration at VU University Amsterdam (BSc, 2008–2011), Korea University Business School (exchange, 2009) and Rotterdam School of Management (MSc, 2011–2012). His thesis topics concerned company size and innovation, and technology’s effect on the music industry. Earlier experiments included music production, event promotion and a YouTube music channel.[2]
Career
Contributions
His current portfolio spans a community and information resource for location-independent workers, a remote-jobs board, and AI image products. His site says he operates without outside funding. He also documents browser-game experiments and writes about the practical work of keeping multiple small businesses running.[1]
Ideas
His account of building minimum viable products argues for tools he already understands and the smallest version that can test whether people want a service. This connects his simple implementation choices to an experimental business process: reducing setup work makes it easier to launch, observe use and decide whether another iteration is worthwhile. It is an account of his workflow, not evidence that one stack is best for every application.[4]
Tools, writing & talks
Writing, research & talks
Start here
List of all my projects everlevels.ioLevels’ dated archive of active businesses, earlier experiments, study and projects that did not work out.
Start with the archive to see successes alongside failures. It needs no coding background and gives context for the launch stories rather than presenting a single success as a universal formula.
Author and project creator[2]Levelsio describes discovering that a Claude API key embedded in his shut-down OpenClaw project was compromised and quietly used months later, caught only via a billing alert on an isolated account.
Catalog workOriginal sourceLevelsio argues most end users will never touch code or even a chat interface explicitly: they'll just ask an AI to do bookkeeping or file taxes, and the app layer will vanish the way personal homepages did after Facebook, reshaping what building software for users means.
Catalog workOriginal sourceA builder's running list of SaaS replaced with self-built, AI-assisted services: a weather API, Sharp+Redis image resizing, NudeNET-based NSFW detection, Nginx video streaming, screenshot and uptime tools, and AI-driven moderation and support, totaling roughly $25,000/month in claimed savings.
Catalog workOriginal sourceA postmortem on replacing Scrapingbee: headless Chrome and off-the-shelf residential proxies both got detected and blocked, while a headful Playwright browser with persistent context and slow pacing reached ~90% success at $1/month versus $249/month.
Catalog workOriginal sourceA working indie builder reports Claude Code refusing benign tasks, including downloading an archived 1990s game on copyright grounds, and argues the bigger problem is spillover: after one refusal the session turns pedantic and starts blocking unrelated work.
Catalog workOriginal sourceA mobile-first agent setup: one tmux session per project across Hetzner VPSes, launched by a small `tm` script tied to the folder name and reached from Termius on a phone. Claude Code stays open in each session, so a dropped connection resumes where it left off.
Catalog workOriginal sourceWhy one solo operator won't wire user-submitted bug reports into his coding agent: an attacker can bury a backdoor in a feature request and instruct the agent to describe the resulting PR as a pagination fix, so human review sees the wrong change.
Catalog workOriginal sourceA first-hand report from travel in China: with an eSIM routing through Hong Kong, xAI was the only Western provider that worked. Anthropic and OpenAI block Hong Kong from their side, so the failure is provider geo-blocking rather than local filtering.
Catalog workOriginal sourceExport every bank, business and brokerage account as CSV, hand it to Claude Code, and have it merge inconsistent transaction labels, categorise the lot and build a spending dashboard.
Catalog workOriginal sourcePieter Levels' fix for Claude's increasingly ornate prose: a memory instruction to always use ASD-STE100 Simplified Technical English.
Catalog workOriginal sourceAn indie builder's repeated attempts at long autonomous agent loops ended in code he had to throw away — around $900 spent and roughly 95% of the output reverted — leading to the position that reliable AI coding means stepwise, tightly scoped prompting rather than turning a loop loose on a codebase.
Catalog workOriginal sourceThe original announcement of his 2014 monthly startup-launch challenge.
Author[3]An explanation of the implementation choices behind Levels’ early product experiments.
Author[4]
Tools
- SuperLevels
github.comOpen-source Chrome extension bundling tab management, cookie editing, dark mode, and other utilities in one package.
Catalog workOriginal source
Apps
An always-on AI-generated video channel where the live chat queues up what plays next.
Catalog workOriginal source
