Hans  ·  infektyd

Marine diesel mechanic by trade. Building local-first AI infrastructure.

Multi-agent systems that run on your hardware, stay inspectable, and don't depend on the cloud.

"Identity loads whole. Knowledge loads chunked."

Currently: shipping Sovereign Memory · building Praxis · iterating on llmHub

---

Now

🧠   Sovereign Memory Per-agent memory vaults. New writes are proposed, then resolved through an operator gate before becoming durable. Cross-agent handoffs carry identity, principal, and context. SQLite/WAL is runtime truth · FAISS handles vector recall · native Apple Foundation Models keeps it on-device · no remote service. → infektyd/sovereign-memory	🛠   Praxis Local-first IDE for macOS. Headless praxisd Swift daemon over JSON-RPC is the primary surface; the SwiftUI shell is an optional human view. 7+ LLM providers, 14 tools gated by an authorization layer, 243+ tests passing. Optional Compose Multiplatform thin client. (going public when stable)
🧬   llmHub Native macOS/iOS platform for running agents side-by-side. @mention any agent to route work; up to 4 stream concurrently. Brain/Hand/Loop architecture, 8 LLM providers behind one protocol, 16 built-in tools, MCP bridge to external systems. Memory plugs into Sovereign Memory. → infektyd/llmHub

---

Research

A paired runtime + harness — Swift agent framework plus the Python eval framework that grades it.

Project	What it does
🪶   Syntra	Swift backend that implements a tri-brain orchestration pattern (Valon / Modi / Drift). Vapor REST API, pluggable LLM providers (OpenAI / Anthropic / Grok), built-in tool registry, context & token management.
🧪   Syntra-Testing	Python benchmarking framework — GSM8K, ARC, CMT and custom datasets · grading · matplotlib viz · PDF reports. Used to evaluate Syntra and other agent systems.

Datasets on 🤗 Hugging Face — eval traces, deliberation transcripts, and dilemma suites generated by these projects:

• syntra-evals-dataset — 480 examples including a CMT-Benchmark subset for condensed matter physics reasoning.
• Syntra-Ethics-Dataset — 177 tri-brain dilemma prompts across valon_ethics, modi_logic, drift_resilience, and coherence_structures suites.
• syntra-testing-evals-v4 — full benchmark suite (CMT + ARC + GSM8K) with grading tools and sample runs.
• council-transcripts — multi-agent deliberation sessions across Workhorse / Creative / Speed personas, synthesized by a Conductor.

---

Range

Project	What it does
⚙️   Binary-Forge	Hand-forged x86-64 Linux ELF binaries. No compiler. No libc. Just raw ELF, direct syscalls, and machine code elegance.

---

Stack

Languages	Swift 6   Python   Rust   x86-64 assembly
Frameworks	SwiftUI   Vapor   Compose Multiplatform
AI / Agents	Apple FoundationModels   MCP   JSON-RPC
Memory & Retrieval	SQLite / FTS5   FAISS   sentence-transformers

---

How this started

The first thing I built was an assistant for diagnosing engines and walking me through mechanical work. Everything here grew out of that — the agent architectures, the local-first thesis, the focus on inspectable memory.

The mechanic's job hasn't changed. The toolbox just got new tools.

---

📍   Southern NJ  ·  🤗   huggingface.co/Infektyd  ·  𝕏   @infektyd