AI in production
Multi-agent systems, evaluation harnesses, and structured-output pipelines running against real users and real data. Models change every few months, so we build the evals, guardrails, and tracing that let you swap them safely.
Senior engineers. Complete systems
A studio of senior engineers shipping AI-native production systems across fintech, healthtech, civic platforms and web3.
Bring us a hard problemClick or tap anywhere to take it apart and rebuild it.
What we build
AI-native engineering across the whole stack: agents and evals first, then the backend, mobile, and chain systems around them. We pick problems where deep engineering judgement actually matters.
Agents & AI platforms
Multi-agent workflows, tool-using agents, and role-aware reasoning pipelines that run against real users and real data.
Evals & LLM infrastructure
Evaluation harnesses, structured-output schemas, model routing, guardrails, and tracing on every prompt and tool call.
Backend systems at scale
Microservices, ledgers, schedulers, and async pipelines that handle real volume, with AI features built in rather than bolted on.
Mobile apps
React Native and native iOS / Android. AI-assisted consumer products, super-apps, and wallets at scale.
SDKs & developer tools
Unity, Swift, TypeScript, and Rust SDKs. Typed APIs and tooling built for developers and agents alike.
Healthtech & fintech
Compliance-aware backends for hospitals, banks, and rewards networks. Ledgers, points, and patient data.
Blockchain infrastructure
Layer 1s, rollups, indexers, and MPC key management. XRPL, Base, EVM, and WASM smart contracts.
Where we've shipped
Years of production work across regulated and high-stakes domains. Clients and projects are under NDA. The stack and the compliance posture are not.
Agentic products, assessment and credentialing platforms, multi-agent workflows, and structured evaluation pipelines.
- multi-agent
- evals
- structured output
- tool use
What we bring
Specific client work stays confidential. The stack, the compliance posture, and the way we operate do not.
Regulated environments
Our engineers have shipped production systems under SOC 2, HIPAA, PCI-DSS, and KYC / AML regimes. Audit trails, access controls, and data segregation get designed in, not bolted on.
Cross-stack depth
React Native and native mobile. Rust, Swift, TypeScript, and Unity SDKs. Microservices, ledgers, schedulers, and indexers. The same engineers spec, build, and operate the system end-to-end.
Independent studio
No bench, no farm-outs, no junior teams behind a senior sales pitch. The engineers you meet are the engineers who ship. We stay on long enough to hand off cleanly, not a day longer.
Blockchain ecosystems
Protocol-level contributions, MPC wallet SDKs, indexer infrastructure, and on-chain provenance. Hands-on across XRPL, Base, EVM, and WASM smart-contract runtimes.
Why 108
The math
The distance from Earth to the Sun is roughly 108 times the Sun's diameter. The distance from Earth to the Moon is roughly 108 times the Moon's diameter. Two unrelated bodies, the same ratio. The number predates us by a few billion years.
The practice
A 108-bead prayer string runs through Hindu, Buddhist, Jain, and Sikh practice. One full cycle of attention, then you start again. Two thousand years of using 108 as a unit of completeness.
000 / 108Run your cursor across the beads.
Hence the name.
We ship complete systems, not prototypes.
How we work
- 01
Spec
We start with the architecture. Threat models, data flow, eval criteria, integration boundaries, written down before code is written.
- 02
Build
Senior teams. No handoffs, no bloat. The engineers who spec it, build it, across every layer of the stack.
- 03
Ship
Production-ready code with CI/CD, evals, observability, and runbooks. We stay on call until it stabilises.
Have a hard problem?
We're selective about what we take on. If you're putting AI into production, or shipping serious infrastructure in any of our domains, let's talk.