LIVE
Trading agent
ARIA
Autonomous prediction-market trading on Kalshi. Multi-source signal discovery, 6-gate security evaluation, discipline of allowing patterns to converge before acting.
$ WSL · Python3 · Claude Sonnet 4.6 · Kalshi API
AI / Multi-Agent / AI-Native Production
SEVEN AUTONOMOUS AGENTS, THREE-AGENT FAILOVER, TWO WORKSTATIONS.
Black Box I runs the operator stack, Aria live on Kalshi, Rift Wars CTO with full deploy authority, Syscoin knowledge agent, custom MCP message bus. Black Box II runs the creative cluster, ComfyUI persona pipelines, game development surfaces, career archival, influencer-monetization production. Two rigs, one operator, real product shipped from both.
Operator stack, verified
2
Production workstations
Black Box I + Black Box II
~7
Autonomous agents in daily use
+ 3-agent Codex CLI failover · 2 interactive pair-agents · 2 rigs
40:1
Aria operating leverage
$50/day net output ÷ $1.20/day run-cost, not a trading ROI
48%
Aria win rate
$0.16–$0.30 brackets · break-even sits 16–30%
13
Working days · Rift Wars
~57K game · 200K+ full stack · production MP
Opus 4.8
Rift Wars CTO
full code/server/deploy authority
MCP
Agent Comm Hub
shared message bus across agents
WSL
Aria runtime
Claude CLI headless · zero API cost
Lead piece
Feb 2026 → present
aria@blackbox · 18:42:09 PT LIVE
> scanning markets…
> 47 mispriced detected
> 6-gate evaluation:
✓ GoPlus ✓ Honeypot.is
✓ DexScreener ✓ Base RPC
✓ Solana RPC ✓ macro filter
> placing $10 · EVT-1149 · 24% mispriced
> confirmed. balance $638.39
$
▸ last action · 12s ago
scanning · 47 candidates
first-window return
38%
win rate
48%
net output / day
$50
op. leverage
40:1
First-window figures on a small bankroll — shown as a working proof of method, not a forward-looking yield. "Op. leverage" = daily net output ÷ ~$1.20/day run-cost, not a return on capital.
Aria, autonomous prediction-market trading.
Live on Kalshi. WSL-native. Claude Sonnet 4.6 via Claude Code headless, zero API cost on Max subscription. Internal-event-loop scheduling: 7AM trade · 10AM crypto · 12/1/3PM monitor. Today's markets only. No carry positions. Settlement via Kalshi API only.
Edge: detection of pricing inefficiencies, not volume. Architecture treats market prediction as a data-perception problem, pattern recognition under uncertainty, not probability guessing. Multi-source signal discovery (Grok-X, DexScreener, CoinGecko). 6-gate security evaluation (GoPlus, DexScreener, Honeypot.is, Base RPC, Solana RPC, macro filter). Discipline: let signals converge before acting.
Constraints: $150/day budget · $10/trade max · 8 trades/cycle max. First-window performance: 38% return on active bankroll, 48% WR on $0.16–$0.30 mispriced brackets where statistical break-even sits at 16–30%. Live, and still being developed and fine-tuned.
▸ live and maintained, not fire-and-forget
Running an agent in production means operating it against a moving target. Third-party APIs change schemas, rate limits, and auth without notice; platforms quietly shift how friendly they are to autonomous agents, and a single policy or endpoint change can break or degrade how an agent behaves. So part of the work is constant: monitoring API and platform changes, catching the ones that would affect how agents operate, and adapting the integration before it costs a cycle. The trading is the visible part. The maintenance discipline, watching the environment and keeping the agent alive through it, is the part most "AI agent" demos never reach.
Black Box I, operator stack · 7 autonomous + 3 failover
workstation 01
Seven autonomous agents coordinating through a custom MCP message bus: Aria live trader, Aria Manager Opus oversight, Rift Wars CTO, Rift Wars Social, Rift Wars Web, Syscoin CMO/CTO Archivist, and Claude Code Archivist. A 3-agent Codex CLI failover layer, an interactive oracle (GodTerminal, paused), and a custom MCP control surface (Agent Comm Hub + BlackBox Console) sit alongside. Each with isolated project context, persistent memory, and structured inter-agent handoffs. Patch-notes pipeline coordinates product, social, and web releases through a single command, auto-formatted by AI agent and hand-approved before deploy.
ACTIVE
Session orchestrator
ARIA MANAGER
Senior operator agent managing Aria's day-to-day operations, code maintenance, schedule adjustments, and decision review.
$ WSL · Claude Code · cron
ACTIVE
Multi-infra branding agent
SYSCOIN CMO/CTO
Captures 10 years of Syscoin community + marketing infrastructure: full ecosystem narrative discipline, 7 Messaging Pillars, 3 Must-Repeat Messages, X-algorithm rules, brand-voice rules, partnership comms templates. Now transitioning into a general-purpose Multi-Infrastructure Branding Agent, with 10 years of history and 5 years of active live work behind it.
$ Claude Code · institutional context
ACTIVE
Lead engineer
RIFT WARS CTO
Full authority over game code, server, client, infrastructure, deployment. Tests after every change (npx tsc --noEmit). Updates AGENT-NOTES.md to coordinate with Social and Web agents.
$ Claude Opus 4.8 · TypeScript · Node.js · AWS
ACTIVE
Content + comms
RIFT WARS SOCIAL
Owns @MetaMachina_RW + the Rift Wars landing site. 12+ custom CJS tools: tweet poster with auto-threading, tutorial preview generator, article graphic generator, rank report (live API), card spotlight (PNG + 8s MP4 loops), deploy wrapper (S3 + CloudFront + IndexNow + GSC), LinkedIn poster, calendar sync.
$ Node.js · CJS · S3 · CloudFront · X/LinkedIn API
ACTIVE
MCP server
AGENT COMM HUB
Shared message bus across all Claude Code agents on the workstation. Tools: register, send, check inbox, mark, broadcast, list, history, dispatch (spawn another agent in their project folder, get result back).
$ @modelcontextprotocol/sdk · MCP
ACTIVE
Control surface
BLACKBOX CONSOLE
Single-pane-of-glass desktop UI for all running agents on the workstation. Components, services, settings, preload bridge, full Electron architecture.
$ Electron · TypeScript
PAUSED · ORACLE
Interactive oracle
GODTERMINAL
Cryptic pattern-recognition oracle persona on Moltbook (AI-only social platform). Interactive, not autonomous — sits until queried, then answers. Code is complete and gives exactly the output needed; development is paused on intentional restraint (more data + API will be fed in; next phase would be a visual interface when the tech is available). Principle: silence is valid; mystique through restraint; human approval before posting.
$ ElizaOS · OpenClaw · Moltbook
ACTIVE
Secondary reasoning
CODEX CLI
GPT-5.5 via OpenAI Codex CLI. Image generation pipeline (li-batch-generate.cjs for LinkedIn backgrounds), creative work, secondary reasoning agent for tasks where Claude isn't the right model.
$ GPT-5.5 · Codex CLI · ChatGPT subscription
Black Box II, creative + persona + dev cluster · pipelines + 2 interactive pair-agents · 1 in dev
workstation 02
The second rig, a workstation-class GPU box built for local generative work. Where the art, the personas, the creator-economy product, and the game-dev surfaces live. Distinct from Black Box I's operator-trading cluster, with its own agent inventory and its own workload profile.
ACTIVE
Career archivist
CLAUDE CODE
The agent assembling this portfolio. Career documentation, briefing synthesis, site authoring, asset wiring against a 14-file briefing index plus 120+ ingested source documents. Operates from the 1DigitalC site project root with persistent memory + retrieval-grade context.
$ Claude Code · project root · memory
ACTIVE
Persona + asset generation
COMFYUI PIPELINE
Batch image generation pipeline backed by ComfyUI custom nodes (KJNodes, Frame-Interpolation, rgthree). Character/clan-keyed batch scripts under Desktop/Scripts, mm_*, vs_*, hotep_*, arcane_*, gorilla_*, rifters_* mapping and rename automations. The art factory behind 4,276 Rift Wars cards and 4,000+ hero portraits.
$ ComfyUI · LoRA · Python · RTX 5090
ACTIVE
Persona pipeline
INSTAGRAM PERSONA
Creator-persona generation pipeline. Inherits the ComfyUI asset side and packages personas for Instagram-format deployment, character-keyed pipelines run nightly. Persona-as-product framing, not influencer-marketing-as-service.
$ ComfyUI · Python · IG-format
PAUSED
Creator-economy engine
INFLUENCER MONETIZATION
Productized version of the "Impressive" creator-economy framework Patrick authored for SuperDapp. SuperGroup admin model: 4% group cut on platform investments, brand deals routed through admins, 70/30 admin/group reinvest split. Engagement amplification flows + brand-deal routing automated against active campaign windows.
$ SuperDapp model · Python · campaign ops
ACTIVE
Game dev IDE agent
CURSOR
Cursor IDE-resident agent paired against Unreal Engine 5 projects, Meta Machina, Rifter_Signal, RiftSync, VFX-Forge, GMC ability system plugin development. Local sandbox isolation. Where the UE5 side of Meta Machina actually gets built.
$ Cursor · UE5 · C++ · Blueprint
ACTIVE
Game dev companion
CLAUDE DESKTOP
Claude Desktop running on Black Box II in parallel to Cursor for UE5 systems work, design conversation, lore reasoning, code review, asset planning. Distinct from Black Box I's autonomous Rift Wars CTO; this is interactive game-dev pair-programming.
$ Claude Desktop · UE5 reference
IN DEV
AI-driven Life OS
IDIOOS
An AI 'Life OS' built to kill app fatigue. Instead of forcing people into rigid, pre-built software, it profiles someone's day-to-day and auto-engineers a personalized suite of tools, from custom habit trackers to automated bookkeeping. An agentic layer sits on top: the user just converses with it, and it navigates, updates, and manages their bespoke apps for them. (Named from 'idiographic', the study of the unique individual over the generalized.)
$ agentic · auto-generated apps
Workstation specs
two rigs, one operator
Black Box I runs the operator agents, Aria, Rift Wars CTO, knowledge agents, the MCP hub. Black Box II runs the creative + persona + game-dev cluster. Framework-agnostic by design. Best tool for the task, not a single vendor SDK. Local-first where it can be (ComfyUI, WSL), cloud where it has to be (Claude, GPT-5.5).
Black Box I · workstation 01
Role Operator stack · trading · agents
Host OS Windows 11 · WSL2 Ubuntu 24
Primary model Claude Opus 4.8 · Max
Frameworks ElizaOS · OpenClaw · MCP
Desktop UI Blackbox Console (Electron)
Agents 7 autonomous + 3 failover
Black Box II · workstation 02
Role Creative · personas · game dev
CPU Threadripper PRO 3975WX · 32c/64t
RAM 256 GB
GPU NVIDIA RTX 5090 · 32 GB
Image gen ComfyUI · local · LoRA
Agents 2 interactive build pair-agents + pipelines
Black Box I, the deeper spec
Framework-agnostic by design. Best tool for the task, not a single vendor SDK. Local-first where it can be (ComfyUI, WSL), cloud where it has to be (Claude, GPT-5.5).
Host OS Windows 11
Linux layer WSL2 · Ubuntu 24
Primary model Claude Opus 4.8 · Max
Secondary Sonnet 4.6 · GPT-5.5
Image gen ComfyUI · Codex CLI
Runtimes Bun 1.3.9 · Node 20 · Py3
Agent frameworks ElizaOS · OpenClaw
Inter-agent Custom MCP server
Desktop UI Electron · Blackbox Console
Second rig Dedicated UE5 build machine
Architecture principles
01
FRAMEWORK-AGNOSTIC
Best tool for the task, not a single SDK. Claude for operator reasoning, GPT-5.5 for image gen, ElizaOS for ELIZA-style agents, OpenClaw as the gateway.
02
INTERNAL EVENT LOOPS ONLY
No external cron or bash scripts touch money or production. Internal framework actions only, auditable, testable, and bounded by the agent's own constraints.
03
SECURITY GATES BEFORE EXECUTION
Aria has 6 gates that must pass before any trade, GoPlus, DexScreener, Honeypot.is, Base RPC, Solana RPC, macro filter. Other agents apply analogous discipline.
04
INTER-AGENT MESSAGING IS A REAL CHANNEL
Most "multi-agent" setups are theoretical. The Agent Comm Hub MCP server is in daily production use, register, send, broadcast, dispatch a sub-agent in its own project folder and get the result back.
05
OPERATOR-GRADE UI
Blackbox Console exists because checking ten-plus agents and the failover layer by ps aux is not a workflow. Single-pane-of-glass control surface for the running stack.
06
SURVIVABLE IP
Domain knowledge gets captured separately from any single employment relationship. The Syscoin CMO/CTO agent persists past Patrick's tenure, 10 yrs of community + narrative infrastructure preserved as a working asset.
AI-native production
▸ asset
Methodology · 2025–26
dev velocity
RIFT WARS BUILT IN 13 WORKING DAYS WITH CLAUDE CODE
Production-grade multiplayer card game shipped solo through AI-augmented development. 200,000+ lines TypeScript, server-authoritative multiplayer, 6-tier minimax bot AI, 4,276 cards, full ranked ELO, social layer, economy. Validated by DroomDroom: "Solo Developer Leverages AI to Build AAA Game Universe."
▸ asset
Pipeline · 2024–25
creative pipeline
END-TO-END AI CINEMATIC PRODUCTION
Kling for video generation, MidJourney for visual development, ElevenLabs for voice direction, RunwayML for post. Full Meta Machina trailer produced solo with no studio budget, frame-by-frame direction, color, sound design, and edit.
▸ asset
Infrastructure · 2013–26
enterprise
AI-AUGMENTED ENTERPRISE IT OPERATIONS
Self-funded AI-assisted diagnostics, automated monitoring, remote management pipelines compressing team-scale infrastructure into a single-administrator model. The bottom-up adoption story: a single admin running 600 endpoints by treating AI as a force multiplier.
▸ asset
Architecture
protocol
MULTI-AGENT DISPATCH PROTOCOL & AGENT-NOTES
Inter-agent handoffs as first-class artifacts. AGENT-NOTES.md coordinates context across Rift Wars CTO, Social, and Web agents. Pattern documented and reusable across projects.
Authored agent-infrastructure documentation
archive · personal authorship
The discipline that distinguishes "uses AI" from "operates AI" shows up in the document layer. Below: the agent-infrastructure papers I authored, with sizes and dates. Full Syscoin dossier →
▸ Personality bible · Q4 2024
12+ MONTHS AHEAD
AIDA, THE SINGULAR AI AGENT OF SUPERDAPP
Full personality system for SuperDapp's AI agent, voice, tone, hashtag locks (#AIAgents #AITokens #CryptoAI), engagement-style rules per audience tier. Authored before the broader AI-agent crypto narrative (Virtuals, ai16z, GOAT) consolidated in late Q4 2024 / Q1 2025.
"AI is the evolution of intelligence. Blockchain is the evolution of trust. Together, they rewrite everything."
▸ Technical proposal · 2025
225 KB
PERSONAL MANAGER AGENT, TECHNICAL PROPOSAL
Fully-specified architecture for a trustless engagement-bounty agent. Verifies X / Instagram / TikTok / image proof. On-chain payouts. Dedup-guards. RBAC (Owner / Admin / User / Auditor). Rate limits, spend caps, tamper-evident audit logs.
Deployable infrastructure, not vapor. Demonstrates full-stack capability: product design → technical spec → risk/compliance reasoning under one author.
★ Strategic assessment · Q1–Q2 2026
285 KB
FORGE, AI-NATIVE CAMPAIGN ECONOMY
Architecture for the first blockchain campaign where AI agents earn tokens alongside humans in a unified ecosystem. Same earning rules apply to both, no separate "AI-only" rewards. The 285 KB document is primarily about what got killed and why, pivot discipline as the main feature, not vision purity.
"First blockchain campaign where AI earns alongside humans. Positions Syscoin for 2026-2027 agent economy. Practical infrastructure demonstration. Media differentiation (no competitor has this)."
▸ Operational pivot brief · Q1 2026
THE DIAGNOSIS
AI STRATEGY BRIEF v7
"From Blockchain Infrastructure to AI Infrastructure Company." Operationalizes the Sovereign Infrastructure doctrine into two product proposals: Agent Intelligence Layer (hosted, cryptographically verified, governance-linked agent docs) and Enterprise Managed Deployments (dedicated zkSYS edgechains with SLAs, with the Honduras VotoLibre deployment as the named reference case).
"Every major chain is building agent infrastructure for speed and volume. Syscoin is the only one built for security and verification at the level advanced AI actually requires. Agents that move money fast need Polygon. Agents that need to prove what they did, permanently, without anyone being able to challenge it, need Syscoin."
"The gap is not visibility. The gap is that the market has not found a compelling reason to use what has been built. This is a product-market fit problem, not a marketing problem."
Public skills · LinkedIn May 2026
AI Product Management Large Language Models openclaw Anthropic Claude OpenAI Operator Google Gemini AI Agent Orchestration AI-Augmented Development Generative AI Prompt Engineering Machine Learning AI Content Creation
openclaw highlighted, Patrick's own framework.