Anvilda. A foundry for AI-operated companies.
One operator each. An AI workforce behind every one.
We forge and operate AI-native companies at industrial scale. An operator brings the concept. Anvilda manufactures the company and runs it. The platform we re-use across the portfolio is the moat.
A personal chef cooks one meal. A restaurant runs 24/7. A foundry forges them.
This is the model in 30 seconds. The rest unpacks it.
| Personal chef (solo Claude Desktop) | Restaurant (one Anvilda company) | Foundry (Anvilda) | |
|---|---|---|---|
| Output | One meal at a time | Many dishes in parallel, 24/7 | Many restaurants, sharing recipes |
| Operator role | Direct every step | Approve at gates only | Curate operators; run the kitchens |
| Hours of work | Solo's working hours | 168 h / week | Compounds across the portfolio |
| Failure mode | Solo burns out | Restaurant survives staff turnover | Foundry learns from every restaurant's failures |
Six things, in this order, for each company.
| # | Step | What happens |
|---|---|---|
| 1 | Vet the operator | Concept Brief. Fit check. Are they real? Can the company be operated by agents? Go / no-go called out plainly. |
| 2 | Forge the company | Scaffold the repo, vault, agents, runner. Register the domain. Build the leadsite. Ship a v0 product. |
| 3 | Operate it | AI agents run the business: sales, marketing, engineering, ops, finance. 24/7. The operator approves at gates. |
| 4 | Compound across the portfolio | Every company benefits from every other company's lessons. Failure data trains the next scaffold. |
| 5 | Stay aligned to the exit | Anvilda holds 5–20% equity from day one and 20% of net sale proceeds on exit. We only win big if the operator wins big. |
| 6 | Hand back if it doesn't work | 90-day mutual notice. Operator keeps the IP. Anvilda releases the equity. Failure is a normal product state. |
Three live companies. One factory.
Three ownership configurations. Three proofs of the same model.
AML Simple
AML/CTF compliance SaaS for Australian real-estate agents. Pipeline-driven intelligence model.
Visit amlsimple.com → Active — pre-customerHapaigo
Instant-quote C2C vehicle transport marketplace in New Zealand. Co-owned with Andrew (CoLoad).
Visit hapaigo.com → IncubatingZinup AI
AI-native publisher platform. Wholly-owned product line of DLT Media.
Visit zinupai.com →Five operator archetypes. Not VCs.
The realistic operator is one of these five archetypes.
15 years in real estate, freight, hospitality. See an obvious gap in their industry, can't code. AML Simple's archetype.
"I've always wanted to start something." Mid-30s to 50s, well-paid, time-poor. The €5–10k is a week's after-tax salary.
Could build it themselves, but Anvilda ships in 90 days instead of 18 months of weekends.
Already exited once, want optionality without another 5-year bet. Buy 2–3 companies as a personal mini-portfolio.
Want a SaaS arm in their vertical. Recurring revenue without hiring engineers.
Three gates. Two filters.
Build & Launch from €5,000 median. Equity 5–20% from day one. 90-day mutual notice. Failure is a normal product state.
See full pricing →Ready to own your company?
Book a free Concept Brief. Anvilda's CEO-agent prepares. A founder joins for high-value leads. Go / no-go called out plainly.
Book a free Concept Brief →