Elastcode is the company. Elastbiz is the product family. v1.0 of Elastbiz Code launches this week — a self-hosted dev platform where AI workers do the building and humans set policy. We're talking to investors who see why that combination changes the cost curve for business software.
We do that by running AI workers on a proprietary platform we wrote ourselves. The platform is 14 years old and production-tested. The AI orchestration is new and shipping. The unit economics — once running — are different from every SaaS and every dev shop because the marginal cost per app trends toward the cost of LLM tokens, not engineering salaries.
Self-hosted git platform with an AI-memory layer. Repos, issues, pull requests, AI reviewers, audit log. Free to self-host. Paid tiers add governance: protected branches, required approvals, Governor.
Token-cost ledger (the metric that makes the unit economics real), per-worker certificate auth, Stripe billing automation, MCP tools for external AI agents, Wiki on AIRecords.
Elastbiz Tables (business data), Elastbiz Connect (team chat), Elastbiz Watch (video monitoring), Elastbiz Mail (AI inbox). Same auth, same data layer, same AI substrate.
We're not building one product. We're building the substrate for software that AI writes and AI runs. The roadmap below isn't speculation — most of it is sketched, several pieces are partially built, and the platform they sit on is already running production workloads.
Code, Tables, Connect, Watch — one auth, one data substrate, one shared AI memory. Self-host or cloud. The flywheel: email arrives → AI reads it → rows appear in Tables → customer wants more features → $50–$500 per feature → continuous revenue.
Mail is email + an AI intelligence layer. Medin is the second AI app — spec pending. Both extend the suite. Same install, same governance, same wallet.
Customer describes an app on elastcode.com → AI estimates and builds → customer self-hosts the output. Marketplace of pre-built templates: dental clinic booking, inventory + barcode, HR + payroll, fleet routing. Hundreds of vertical apps, generated on the same substrate that runs the suite.
The operating system robots run on. R for Robots, Runtime, Remembering. The same AI-memory substrate scaled to physical systems with hard real-time constraints. Not 2026 work — but the architecture is being designed now, and the platform we've built is what makes it possible.
Traditional custom-software dev shops bill hours. SaaS companies bill seats. We bill per-app or per-tier with no per-seat creep, and our cost per app trends toward LLM token cost as AI workers absorb more of the implementation.
That means:
The 14-year platform is the moat. The AI orchestration is the new leverage. Together they let us price aggressively and still earn an unusual gross margin.
We're a small team in Cyprus running a long platform. If you understand why a tight human team plus an AI worker roster on a 14-year substrate is a different kind of company — we'd like to talk.
Elastcode Ltd · Cyprus · Founded 2010
