An AI research lab founded in November 2024 in London. Personal intelligence is our central thread — but we work across language models, architectures, reasoning and the systems on top, and we ship the results in public.
Building personal and general intelligence — for every mind.
We treat frontier AI as one connected research programme. Personal intelligence — persistent memory, self-reflection, per-agent cognition — is the through-line that ties our work together, but it's not the only thing we work on. Some of our research is on language models and reflection (AVALON-2B). Some is on byte-level architectures and routing (PLMR). Some is on systems we haven't named yet (Hydra). Some ships as open weights. Some ships as commercial infrastructure (Hypersave) or applied products (Khyaa, the Nuro stack). The pattern: do real research, ship real things, move the frontier a little forward each time.
We build that compounding in three layers — open research, commercial infrastructure, applied products — and we hold ourselves to seven principles so we don't lose the plot along the way. The applied work is not a distraction from the research. It is the test of it.
We measure the lab by what we ship and what we publish — not by headcount or spend.
Memory is the missing layer. Anyone who has watched a frontier model reset between sessions knows the ceiling isn't parameter count — it's that nothing carries forward. The bet on personal intelligence as the compounding substrate for general intelligence is the right bet, and it's startlingly under-served.
Composite quote drawn from advisor conversations. Attribution by request — full named advisory list on /press.
Open-weights papers and models. Apache 2.0 by default. AVALON-2B is live; the PLMR preprint is imminent; Hydra is in active development. The published research is the marketing.
See research →Hypersave — the cognitive memory layer for AI agents. Five sectors, hybrid retrieval, sub-200ms p95, SOC 2 Type II. Beats every adjacent agent-memory system on LoCoMo head-to-head.
Hypersave →Vertical and consumer products that prove the research works. Khyaa, Nuro Studio, Nuro Chat and Nuro One — compliance-aware, citation-backed, on-device first.
Products →Finance, legal, hiring, partnerships and the boring infrastructure that lets the other three layers compound. Small, deliberate, and built for a decade rather than a quarter.
Careers →Tap any principle to read the long form. The first three are the load-bearing ones.
Our research output is open by default — weights, training code, and the synthetic-data pipelines that produced them. AVALON-2B is the proof. Our products are commercial: Hypersave, Khyaa and the Nuro stack are how we fund the lab. The split is deliberate, and it's how Anthropic, OpenAI, Mistral and DeepMind all actually operate.
Every advance we ship has to make at least one specific mind better — one user, one agent, one team — before it makes the world better. We do not chase abstract benchmarks at the cost of the unit we serve.
We ship in public, incrementally. We do not pre-announce. We do not promise products we have not built. The published research is the marketing.
Every claim is auditable. Every benchmark we publish has reproducible code. Every Khyaa answer cites its source. Every Hypersave query carries a confidence score. We treat citation as a first-class product feature.
A frontier lab that doesn't ship what it builds is a research department. We ship Hypersave, Khyaa and Nuro Studio alongside AVALON and PLMR. The applied work is not a distraction from the research — it is the test of it.
Personal intelligence is not luxury infrastructure. The cognitive memory layer should run on a phone, in a regulated bank, in an SME, in a classroom — not just in a hyperscaler. Geography of compute should not gate access to intelligence.
General intelligence is the goal. We don't think it's three months away, and we don't think the path runs through a single 10-trillion-parameter pretraining run. We think it runs through compounding — through architecture, through memory, through reflection. We are willing to take a decade.
We are a UK private limited company. Plain-text and auditable — every claim on this page is filed publicly with Companies House.
View Companies House filing ↗Visits by appointment only. The fastest way to reach the lab is email — we read everything sent to hello@.
When a paper drops, when a product ships, when something material happens — that's when you'll hear from us. Unsubscribe anytime.
We hire research engineers, applied engineers and GTM operators on a rolling basis. Small lab, high signal, decade-long horizon.
