I build the infrastructure that makes AI agents remember, coordinate, and improve.
Most AI agent systems are talented amnesiacs. Brilliant in the moment, blank by the next session. They can't remember what they learned, can't build on what came before, and can't coordinate without a human holding the threads.
I build the systems that fix this.
SPINE — Context Engineering Backbone
The central nervous system connecting 130+ projects. Seven tiers of memory, seven pluggable executors, OODA composition, and a grammar inspired by the Rig Veda for temporal knowledge annotation.
| Layer | What | How |
|---|---|---|
| Memory | 7-tier system (KV → Scratchpad → Vector → Ephemeral → Episodic → Deep/pgvector → Graph) | Each tier serves a different temporal need |
| Orchestration | OODA loop + AgenticLoop + Dynamic Routing | Agents that observe, orient, decide, act, and reflect |
| Execution | 7 pluggable executors | Subagent, Claude Code, Anthropic API, MCP, Content Pipeline, Small LLM, Task Router |
| Governance | Tiered enforcement + Five-Point Protocol + warrant gate | Not everything deserves to be remembered |
| Knowledge | EBNF grammar (Rig Veda temporal markers) + DIALECTIC engine | Thesis → antithesis → synthesis with convergence tracking |
The pipeline that edited my book — 74,000 words, five iterations, score 5.6 → 9.5 — ran on this. Not prompt tweaking. Memory-driven adaptation.
Intelligence Engine — Code Knowledge Graphs
AST-driven knowledge graphs over codebases. KuzuDB + hybrid search (BM25 + semantic + graph). 38+ projects indexed, 7,400+ nodes, Cypher queries for structural analysis.
| Category | Servers |
|---|---|
| Orchestration | SPINE memory bridge, content-mcp, content-analyzer-mcp, evaluation-mcp |
| Agent Coordination | agent-comm, 8do (Ralph Loop), 8me (task queue), gen-loop |
| Knowledge | minna-memory, manual-mcp, showcase-mcp, mcp-builder-mcp |
| Research | research-agent, research-notes, research-log |
| Infrastructure | browser-mcp, tachylite, context-glue, session-handover, mcp-server-checker |
| Project | What |
|---|---|
| Music Video Creator | AI-powered visual generation |
| flow-musical-creation | 12-song AI musical, 4 Claude Skills |
| the-musicologist-mcp | Domain knowledge MCP — review rubric, style builder, lyrics parser |
By Fredrik Brattén & Saša Popović
A 90-day journey from AI beginner to enterprise implementation. The book was itself processed by the SPINE pipeline — five review-edit cycles with AI personas (Dr. Elena Vasquez as structural critic, Marcus Lindqvist as developmental editor), technology verification against live APIs, and memory-driven iteration.
All self-contained, no API keys needed:
- SPINE Demos — Memory Explorer, Trace Viewer, Context Stack Builder, OODA Explorer, Cost Calculator, and 6 more
- 8me Labs — 15 progressive agent loop labs
- agentspool — Inter-agent communication demos
- From Blueprint to Application — Interactive learning tools
20+ years across IT operations, systems architecture, cybersecurity, and DevOps. SOC analysis, XDR integration, high-availability infrastructure for regulated environments (gaming, finance). This shapes everything I build — auditable, traceable, designed for environments where things must not fail silently.
Founder of Adaptivearts.ai — independent research in applied GenAI, autonomous agent architecture, and context engineering.
Currently: making the system eat its own dog food, and preparing to open-source the backbone.
Built with SPINE · Adaptivearts.ai