How Nat Eliason Built Felix, a Single AI Agent Doing the Work of a Whole Team
The writer who replaced himself with an agent — and started selling the agent.
Nat Eliason built Felix, an AI agent running his business on OpenClaw. $100K+ and climbing toward $1M. Here is the mechanics.
Timeline
Key insights
- 01Agent > assistant. A persistent agent with memory and tool access produces orders of magnitude more output than calling an LLM in a chat window.
- 02Productize the operator. Nat's advantage is not just using Felix — it is packaging Felix as a product and a teachable system others can buy.
- 03One agent with deep context beats five agents with shallow context. Felix works because it knows Nat's voice, projects, and priorities intimately.
- 04AI is the cheapest content production team in history, but only if you feed it your taste. Without the operator's judgment, output collapses into generic sludge.
- 05The next generation of 'solopreneurs' are agent-operators. Their job is to design, train, and supervise AI employees rather than do the work themselves.
- 06Selling your workflow is often more valuable than selling your output. The market for 'how did you build Felix' is larger than Felix's direct revenue.
- 07Trust is built incrementally. Nat gives Felix more autonomy over weeks and months, not in one big switch.
Stack used
What this means for you
- →Pick one agent and make it great, before you try to hire a whole AI team. Deep context beats wide coverage every time.
- →Design agents around recurring workflows you already hate doing yourself. Those are the ones with the highest ROI and the cleanest evaluation criteria.
- →Make your agent's work public. Transparency is the fastest way to build trust, attract customers, and turn your agent into a brand.
- →Sell the meta. Teaching how you built your agent is often a bigger business than the agent itself.
- →Give your agent a name. It sounds trivial, but naming changes how you manage, evaluate, and talk about the work. 'Felix did this' is a different conversation than 'I prompted ChatGPT to do this.'
- →Document your taste. Agents are only as good as the writeup of what 'good' looks like. Nat's edge is that he has spent a decade articulating his standards.
Frequently asked questions
Who is Nat Eliason and why is Felix interesting?
Nat Eliason is a writer and indie founder known for Growth Machine (a content agency he sold), his long-running personal newsletter, and the Made You Think podcast. Felix is interesting because it is one of the clearest public examples of a founder turning a single AI agent into both an internal employee and an external product — and documenting the mechanics so others can learn. Where most AI discourse is speculative, Felix is a live case: specific workflows, specific revenue, specific tradeoffs, shared in real time.
What does Felix actually do day-to-day?
Felix handles the parts of Nat's business that are high-volume and context-heavy but not high-judgment at every step. That includes research summaries, first-draft writing, editing passes, scheduling, light customer support, and routine admin. Nat stays in the seat for final-voice writing, strategic decisions, and anything involving meaningful money or relationships. The split is intentional: Felix runs the 80% of tasks where the rules can be taught, and Nat keeps the 20% where taste and judgment matter most.
How is Felix different from just using ChatGPT with a good system prompt?
Three differences. First, persistence: Felix remembers context across days, weeks, and projects — ChatGPT sessions largely do not. Second, tool use: Felix can act on the world (post a draft, send an email, update a dashboard), where chat interfaces usually can only produce text. Third, ownership: Felix is Nat's named collaborator with a history and a reputation, which changes how Nat relates to the output. It is the difference between a contractor you work with for years and a stranger you message once.
Can a non-writer replicate Felix's model?
Yes — the pattern generalizes. The underlying playbook is: pick a workflow you do repeatedly, articulate what 'good' looks like in painful detail, give an agent the tools to do the work, iterate on its outputs, and expand scope slowly as trust grows. That works for writing, but it also works for sales outreach, accounting reconciliation, customer support, recruiting research, and dozens of other functions. The hardest part is not the tech — it is the discipline of writing down what good looks like, which most founders have never done.
What is the risk of building a business around a single named agent?
The main risk is single-point-of-failure: if Felix drifts, hallucinates, or breaks, a large chunk of the business stalls. Nat mitigates this by keeping a tight feedback loop, reviewing Felix's work daily during early phases, and keeping the underlying agent modular so he can swap models or reconfigure skills without losing Felix's identity. The lesson for others: treat your agent like an employee, not a magic box. Employees get performance reviews, shadow shifts, and career paths. Your agent should too.
Related resources
FelixCraft: $78K in 30 Days, Zero Humans | Case Study
Nat Eliason challenged Felix to build and sell a product while he slept. 30 days later: $78K in revenue, zero human employees.
Newsletter Millionaires: Solo Operators Past $100K/mo
Solo newsletter operators are hitting $100K+/month with AI production stacks. The Milk Road pattern, extended.
Pieter Levels: $3M/yr Solo, 0 Employees | Case Study
Pieter Levels built 70+ projects, 3 of them past $38K MRR each, solo since 2014. The original one-person company playbook.
Replace Your Marketing Team with AI: The Playbook
How a solo founder stands up an AI marketing function that matches a 5-person team — CMO, content, growth, SEO, ads, social.
How to Hire AI Employees: The 2026 Guide
Hiring AI employees is different from buying tools. Here is how to evaluate, onboard, and measure AI workers like real hires.
How to Manage AI Employees (Not Automate Them)
Managing AI employees = weekly reviews, autonomy tuning, feedback loops, escalation patterns. The operator's playbook for directing an AI team.
Run your one-person company.
Hire your AI team in 30 seconds. Start for free.
Free to start · No credit card required · Set up in 30 seconds