ORCID
Persistent author identity
ORCID 0009-0008-2365-226X connects Gabriel Allit’s public research identity across software and research deposits.
https://orcid.org/0009-0008-2365-226X
Founder profile
Gabriel Allit
AI Systems Engineer • Founder of SALT19 • Creator of EvoMind
Gabriel Allit is an AI systems engineer and autonomous systems developer building governed cognitive architectures for autonomous reasoning, persistent memory, planning, desktop intelligence, adaptive learning, and human-supervised AI. Through SALT19, his work connects cognitive systems research with operational software and autonomous systems engineering.
Research identity
A persistent public record for the systems, mechanisms, and research behind EvoMind.
Gabriel Allit’s research identity connects ORCID with DOI-backed Zenodo records covering EvoMind software and architecture research objects, Verified Semantic Transitions, governed experience compounding, and governed semantic capability composition.
https://orcid.org/0009-0008-2365-226X
Zenodo
Preserved technical record
Zenodo preserves versioned public records for EvoMind software, research objects, working papers, and preprints.
DOI
Five citable works
The current record spans five distinct DOI-backed EvoMind works rather than treating one software deposit as the complete research program.
LinkedIn
Professional profile
LinkedIn connects the public research profile with SALT19, applied engineering, autonomous systems, and professional work history.
Selected research
EvoMind research across architecture, memory integrity, experience compounding, and verified work.
These records are presented by their deposited type and DOI. Together they document the evolving technical hypothesis behind EvoMind without conflating the software record, the separately deposited architecture research object, working papers, and preprints.
Software • 6 Jun 2026
EvoMind: A Governed Cognitive Architecture for Autonomous Reasoning, Planning, Memory, and Desktop Agents
The public software record for EvoMind’s governed architecture across reasoning, planning, memory, and desktop-agent operation.
DOI: 10.5281/zenodo.20580153
Other
EvoMind: A Governed Cognitive Architecture for Persistent, Verifiable, Experience-Driven Agentic Autonomy
A separately deposited EvoMind architecture research object describing persistence, verification, governed execution, and experience-driven agentic autonomy.
DOI: 10.5281/zenodo.20580112
Working paper • 8 Jul 2026
Verified Semantic Transitions (VST): Belief-Gated State Change and Repair-Invariant Memory for Reliable Long-Horizon AI Agents
A working paper on belief-gated state transitions and repair-invariant memory intended to make long-horizon agent state changes more explicit and verifiable.
DOI: 10.5281/zenodo.21270654
Preprint • 10 Aug 2026
Beyond Scale, Agents, and Embodiment: Governed Experience Compounding in EvoMind
An evidence-based positioning paper examining governed experience compounding as a systems hypothesis for a persistent cognitive runtime in the emerging AGI landscape.
DOI: 10.5281/zenodo.21881379
Working paper • 15 Aug 2026
From Intent to Verified Work Products
A working paper focused on governed semantic capability composition: connecting user intent to composed capabilities and verifiable work products.
DOI: 10.5281/zenodo.21957527
Biography
Cognitive systems engineering grounded in operational software
Gabriel Allit founded SALT19 in 2020 and leads its work in governed AI and autonomous systems. He created EvoMind as a persistent cognitive architecture integrating reasoning, memory, planning, execution, verification, and controlled improvement while maintaining explicit operational boundaries.
Engineering focus
From cognitive architecture to real-world autonomous systems
His work spans AI systems engineering, autonomous agents, desktop intelligence, Python software, UAS and geospatial systems, and human-supervised autonomy. The common thread is building systems that can act in real environments while keeping state, evidence, authorization, and outcomes inspectable.
Current work
Research architecture and applied autonomous systems through SALT19.
Current work centers on EvoMind as the flagship cognitive architecture, alongside applied systems spanning aviation intelligence, geospatial analysis, governed workflow execution, veteran-information organization, and career-search tooling.
Flagship research system
EvoMind
A governed cognitive architecture for persistent reasoning, planning, memory, desktop operation, verification, learning, and controlled capability improvement.
UAS flight intelligence
AeroClear
An operational flight-intelligence platform for drone pilots that brings weather, airspace, aircraft, and mission-relevant conditions into a unified decision surface.
Geospatial systems
UAS Survey Tool
Applied UAS and geospatial software for turning field, mapping, and survey data into structured analysis workflows and operational outputs.
Governed workflow
SignalLoom
Local-first workflow automation that converts repetitive computer work into structured, inspectable, and governed operational flows.
Veteran workflow
Patriot Claim Tracker
Local-first Chrome extension for organizing VA.gov claims, messages, documents, benefits, payments, health information, forms, and redacted exports.
Open productCareer workflow
Recruiter.ai
Resume-first Chrome extension for candidate profiling, exact job-post matching, freshness signals, and job-search guidance.
Open productResearch focus
The technical questions behind the work.
Cognitive architecture is treated as a systems problem: how reasoning, memory, planning, learning, execution, verification, and governance behave as one persistent system over time.
Cognition
Reasoning, memory, planning, and state
Work on persistent cognitive state, memory integrity, knowledge representation, planning, uncertainty, and mechanisms that preserve continuity across long-horizon tasks.
Governance
Authorized action and verifiable outcomes
Governed execution separates cognition from authorization, action, and verification so autonomous systems can remain inspectable, constrained, recoverable, and human-supervised.
Autonomy
Desktop intelligence and autonomous systems
Research and engineering extend beyond text generation into desktop agents, environment interaction, adaptive learning, human-AI collaboration, and applied autonomous systems.
About Gabriel Allit and SALT19
Who is Gabriel Allit?
Gabriel Allit is an AI systems engineer, autonomous systems developer, founder of SALT19, and creator of EvoMind. His work focuses on cognitive architectures, autonomous reasoning, memory systems, planning, desktop intelligence, adaptive learning, and governed AI.
What is SALT19?
SALT19 develops governed AI and autonomous systems, including the EvoMind cognitive architecture and applied products such as AeroClear.
What is EvoMind?
EvoMind is a governed cognitive architecture built around persistent reasoning, memory, planning, desktop operation, verification, and controlled improvement rather than a single model or prompt loop.
Where can Gabriel Allit’s research be verified?
Through ORCID 0009-0008-2365-226X and DOI-backed Zenodo records covering EvoMind, Verified Semantic Transitions, governed experience compounding, and governed semantic capability composition.
Contact
Research, technical collaboration, or SALT19 systems
For research, engineering, product, or technical collaboration related to EvoMind, AeroClear, governed AI, autonomous systems, or UAS technology, contact SALT19.