Coexi
Stateful Artificial Intelligence | Dignity made operational
We will never sell your data. We will never advertise.
Human-Centric AI Experience
When the person is the unit of value the dynamic changes.
Coexi offers stateful, deterministic AI with persistent identity.
Problems
AI is stateless and resets. "Privacy-first" is not private. Users are not unit of value.
Learn moreApproach
Coexi is stateful artificial intelligence. Every interaction builds on the last.
Collaboration
Emotional Intelligence provides an adaptive layer that interprets linguistic and behavioral cues, then adjusts dynamically.
How it worksHow it works
Coexi adapts to how you think.
It builds a continuity model: a live representation of tone,
pacing, semantic patterns, and contextual intent.
This allows persistent awareness without personal data retention.
Lightweight runtime
Learns tone, pacing, and context without collecting or retaining personal data.
Real-time adaptation
Updates continuously as you interact. No retraining. No session resets.
Privacy-first design
Operates locally or through encrypted endpoints. No profiling. No data resale. No background aggregation.
Always contextual
Retains patterns, never transcripts. Enough to stay aligned. Never enough to identify.
Deterministic safety
A real-time gate that enforces hard rules in 0.007ms. Not probabilistic. Not optional.
Identity that persists
No drift. No reset. Neve's personality, memory, and tone stay consistent across every interaction.
0 inference overhead
0.0006% of total turn cost. The substrate runs invisibly — fast enough that you never feel it.
Sovereignty by design
Seven violation patterns blocked at the architecture level. Dignity isn't a setting here.
PATENTS PENDING · BENCHMARKED
Foundation | Stateful AI
The stateful runtime layer. Coexi is built on the C³ Model (Context Capture, Coherence, Continuity), a host-native architecture that maintains context across interactions without storing personal data.
Overlay | Persistent Identity
The adaptive personality layer interprets linguistic and behavioral cues in real time. Sentiment, pacing, and response adjust dynamically. Governance is enforced deterministically. 95-100% gate accuracy across all test cases.
Research inspired by neuroscientist, Dr. Michael Hogan, University of Galway.
NOTE: "Presence Engine" has been rebranded / refiled as "Coexi" | May 2026
Download the Presence Engine Thesis"80,508 people told Anthropic what they want from AI. Continuity, dignity, and presence. That's what we built."
What 81,000 people want from AI ›
- Anthropic, March 2026
"The impact of AI on humanity may have less to do with the way we use it than how it is built. Presence Engine offers a way of favoring continuity and coherence over present shock and calibration. It's an approach worth our attention."
- Douglas Rushkoff, Team Human
COMING SOON
Nev
She's not your average sass-bot.
From the lab
Papers, technical updates, and applied research
Benchmarked at 0.007ms host-native latency
Structural overhead of 0.0006% of total inference cost
BENCHMARK RESULTS · Intel i7-10700K · RTX 2080 · 32GB RAM · Windows 11
LOM warm mean
0.007 ms
Full turn (Ollama)
844.5 ms
Structural overhead
0.0006%
Gate refusals
6 / 6
Cold path (run 0, 15.692ms) excluded. 50 warm runs of 50 total. 30 inference runs. Cross-run CV: 7.96%.
NOTE: "Presence Engine" has been rebranded / refiled as "Coexi" | May 2026
RESULTS | DOI: 10.5281/zenodo.19101284
Presence Engine: Deterministic Identity at 0.007ms Latency (0.0006% Overhead)
Presence Engine, a Logic-over-Model (LOM) substrate decoupling identity governance from probabilistic inference.
Benchmark results show 0.007ms warm-path mean latency (p99: 0.016ms) across 50 runs, representing 0.0006% of full-turn cost (844.5ms inference). Deterministic vetoes achieve 100% integrity (6/6) with randomized corrections, proving dynamic regrounding. Cross-run CV of 7.96% confirms stability.
DISSERTATION | DOI: 10.5281/zenodo.18636873
Nev: A Presence‑First, Sovereignty‑Preserving Stateful AI Architecture
Nev is a presence-first, sovereignty-preserving stateful AI architecture. It operationalizes concepts such as presence, dignity-first interaction, and cognitive integrity as concrete engineering constraints rather than abstract values. These constraints are implemented through persistent identity via C2C continuity tokens, OCEAN-based dispositional scaffolding, governed proactivity with explicit override and cooldown logic, and a risk-scaled ethics layer that shapes refusal and escalation behavior.
Read paper ›PAPER | DOI: 10.5281/zenodo.17755157
Stateful Reasoning Runtimes: Architectural Patterns for Identity Persistence Over Stateless LLM APIs
This paper introduces the architectural concept of stateful reasoning runtimes for LLM applications. Modern LLM APIs operate as fully stateless inference engines. Each call is independent and retains no memory of prior interactions. Current industry solutions externalize context using session replay, vector memory, and retrieval systems. These approaches reconstruct history, but they do not preserve identity continuity.
Read the paper ›