The frictionless AI engine.
ZeroVelocity builds the tools that sit between your models and what you're trying to build — model-agnostic, designed to thrive on frontier models, and getting sharper every version.
Capability is racing ahead. The tooling isn't.
Frontier models keep getting more capable. The way most teams use them hasn't kept up — brittle prompts, no memory between sessions, no guardrails, and no clear record of what an agent actually did. The bottleneck stopped being raw capability. It's the layer around it.
Meet Zara.
The family of AI software from ZeroVelocity, built to improve, optimize, and fill the gaps in the AI dev setup you already have.
Zara isn't a model, and it isn't a wrapper. It's the harness, memory, and tooling layer that makes whatever you already run more capable, more consistent, and more yours. Model-agnostic by principle — frontier-native by design.
Model-agnostic, frontier-native
Works with any model. Built to thrive on the best.
Recurrent learning + memory
Gets better with use, and remembers what matters.
Gap-fillers, not rip-and-replace
Improves your stack instead of forcing a migration.
Bunnita — the proof.
It started as a spark. In 2025, Bun was the first attempt at a better harness around a model — enough to prove the direction was worth chasing. Early 2026, that became Bunnita: the serious step into real research, deliberation, and iteration toward the next big step in AI.
Bunnita is a recursive, self-improving harness with persistent memory and a governed build process. It's not a demo — it's on its sixth shipped version, each one measured, documented, and used to design the next.
- 6 shipped versions
- v1.0 (May 2026) → v1.6 (June 2026), v1.7 in progress
- Capability-native core
- A lean architecture designed for frontier models
- Persistent memory graph
- Sessions become long-term signal, not throwaway context
- A governed build loop
- Every version improves on the last, on the record
Version history
- v1.0Foundational architecture — permission tiers, cloud memory graph
- v1.1Personality modes, subagent inter-comms, staleness handling
- v1.2Intent escalation, quarantine, browser automation
- v1.3Recursive feedback loops
- v1.4Split architecture + dynamic extensions
- v1.5Structured execution — tool specs, verification pipeline
- v1.6Capability-native core, scaffold extension, knowledge base
- v1.7Consolidation & proof· in progress
It improves through a team that actually deliberates.
Every build of Bunnita is put to work by a team of specialized agents — each with its own expertise, tools, personality, and memory. They analyze the last version, argue about what's weak, and plan the next. It's not a single pass by a single model. It's a process that compounds.
When a frontier model first analyzed Bunnita, it was blunt: the real value here is the process, not the prompt. So we handed it the keys and let it build a version alone. Reviewing its own work, it reversed — the worst gaps were all at the seams no single author checks. Its verdict: the committee "isn't bureaucracy; on this release it was the missing safety check." We added two specialists and kept going.
Vision proposes; verification proves. That balance is the point — and it's why the tools get sharper instead of drifting.
The team
- IsabellaSpecialist
Safety & behavioral architecture
- DanielSpecialist
Tools & capability architecture
- PriyaSpecialist
Memory & persistence architecture
- MarcusSpecialist
Tone, formatting & UX
- SørenJoined after v1.6
Engineering architecture & model expertise
- RenataJoined after v1.6
Integration verification & empirical rigor
- ZachariConvener
Convener, final authority
What's next.
ZeroVelocity is a company, not a single product. Here's where the Zara family is headed.
A new CLI
In designA command-line tool and suite, rebuilt from the approach up for frontier models and what comes after.
Ironclad guardrails
ComingDeterministic boundaries that keep agents exactly in scope, by construction.
Skill auditing & management
ComingReviewable logs and lifecycle control for every tool call and skill.
Cost & token optimization
ComingRouting and lean prompting that cut token spend without cutting logic.
Workflow portability
ComingAgents, skills, and workflows as portable, packaged units.
Bunnita v2.0
Late 2026The first public release, shipping with tools and skills.
Built by someone who's shipped at the edge for 30 years.
ZeroVelocity is founded on three decades of coding, startups, and bleeding-edge technology — the kind of track record it takes to build the hard layer and see where AI is actually going. Founded by Zachari, ZeroVelocity is architecture-first: designed for the models of today and the ones that haven't shipped yet.
Bunnita v2.0 lands late 2026.
Get on the list. Be first to the CLI, the tools, and the release.