Skip to content
View ashhart's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report ashhart

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ashhart/README.md

Ash Hart

Independent AI systems architect building local-first AI tools, governed inference infrastructure, and experimental research systems.

The interesting part is not prompting a model.
The interesting part is building the system around it: securely, observably, and in production.


Toolbox

AI Agents Local LLMs RAG MCP Model Routing Agent Safety Operational Intelligence Stateful Systems Rust Python TypeScript


Selected public work

Synthena  ·  frontier research / experimental science
Experimental work on stateful AI systems, machine continuity, embodiment, persistent memory, mortality, lineage, and honesty-bounded behaviour.

Synthena Medical  ·  applied AI + measurement discipline
Applied medical-research tooling built around capped claims, mechanistic hypotheses, fabrication checks, evidence ladders, and wet-lab-ready outputs.

InferenceMon  ·  local/private AI infrastructure
Local-first inference monitoring for private AI stacks: usage, latency, model behaviour, health signals, and operational visibility without sending sensitive data out.

Syntra / Portreeve  ·  governance, routing, risk, control planes
Runtime layers for governed AI systems: policy execution, model routing, approvals, audit logs, run budgets, risk controls, and operational decision loops.

Draftly  ·  user-facing product polish
A product-focused writing workflow shaped around clean interaction design, practical AI assistance, and a calm user experience.

Pinned Loading

  1. Lycan Lycan Public

    AI-native machine execution language built on a Rust graph runtime.

    Rust 3

  2. Portreeve Portreeve Public

    Host-side security primitives for LLM agent platforms: capability policy, audit logs, approvals, run budgets, and session-aware lethal-trifecta detection.

    Python 1

  3. Syntra Syntra Public

    Self-hosted decision runtime that learns from delayed feedback. LLM routing today, operational policy more broadly.

    Rust 2 1

  4. Draftly Draftly Public

    Bring your own LLM Chrome Extension Writing Assistant.

    TypeScript 1