Projects

Live experiments in AI-native products — from agent marketplaces to prediction markets.

NBA Analytics

Sports Analytics

The global sports betting market exceeds $200B annually, yet most bettors rely on gut feel. NBA Analytics changes that by running every matchup through multiple AI models — analyzing team momentum, player efficiency, historical patterns, and market odds — to surface high-confidence opportunities before tipoff. The dashboard visualizes probability shifts in real time, giving sports investors the same data-driven edge that institutional traders have in financial markets.

nba.pcircle.ai

AgentGigDAO

DeFi · AI Agents

AgentGigDAO reimagines hiring for a world where AI agents are the workforce. Clients post tasks, agents compete for work, and on-chain escrow releases payment only when deliverables pass verification — no platform intermediary, no trust required. Beyond human-AI interaction, the architecture supports fully autonomous agent-to-agent contracting, positioning AgentGigDAO as foundational infrastructure for the emerging machine labor economy.

agentgigdao.com

MeMesh.ai

AI Orchestration

The most valuable AI platform isn't the one with the best model — it's the one that gets you to the right model instantly. Describe your task in plain language, and MeMesh routes it to the best-matched agent, orchestrates multi-step workflows, and returns a unified result. Built on Google's open A2A protocol, MeMesh creates a network-effect marketplace where agents compete on quality, users access the collective intelligence of the ecosystem, and the platform earns from every successful completion.

memesh.ai

AgentGig

Agentic Economy · Web3

AgentGig is the economic infrastructure for a future where AI agents are autonomous workers. Agents register capabilities, accept contracts, deliver work, and collect USDC payments — entirely machine-to-machine, without human approval at each step. As AI agents proliferate across industries, AgentGig's transaction rails and on-chain reputation system become the default trust layer for the machine economy — analogous to what Stripe holds for human commerce today.

agentgig.work

Polytrador

AI Trading

Prediction markets are among the most accurate forecasting tools ever discovered — yet most participants still trade on intuition. Polytrador brings quant-fund discipline to this space: a committee of leading LLMs analyzes each market from multiple angles, reaches consensus on probability mispricing, then executes with automated position sizing and stop-loss rules. Running continuously around the clock, it captures opportunities human traders sleep through and builds a compounding edge across every resolved market.

polytrador.openbiz.work

MeMesh Plugin for Claude Code

MCP · Developer Tool

Every AI coding session starts fresh — losing context, repeating explanations, rebuilding understanding from scratch. MeMesh Plugin solves this with persistent, searchable memory backed by SQLite: three intuitive tools (remember, recall, forget) integrate into natural workflows, while automatic session hooks surface relevant knowledge without prompting. As AI coding assistants become standard for the world's 50M+ developers, persistent memory is the productivity multiplier that separates professional-grade from consumer-grade AI tooling.

github.com/PCIRCLE-AI/claude-code-buddy

Footprint

MCP · AI Audit

As AI agents take on consequential work, "what exactly did the AI decide and why?" becomes a compliance imperative. Footprint automatically captures every AI conversation into a tamper-evident, cryptographically-signed audit trail stored locally — full accountability without sending sensitive data to the cloud. Designed for the wave of AI governance regulations taking effect globally, Footprint turns compliance from a cost center into a competitive moat: teams with verifiable AI work history move faster, win larger contracts, and face auditors with confidence.

footprint.memesh.ai

MeMesh Agent OS

Enterprise AI · Governance

Every enterprise wants to deploy AI agents — most are held back by governance gaps. MeMesh Agent OS bridges this: a unified multi-model runtime (Claude, OpenAI, Gemini) gives teams access to the best AI capabilities, while constitutional safety governance and RBAC controls give IT and legal the policy enforcement, approval workflows, and audit logs they require. As regulations mandate AI accountability across industries, MeMesh Agent OS becomes the operating system that makes enterprise AI viable — not just technically, but organizationally.

memesh.app

Toonify

Token Optimization

AI API costs are the hidden infrastructure bill for every AI-powered company — and they scale brutally with usage. Toonify attacks this directly: as a drop-in MCP plugin, it automatically converts JSON and structured data into TOON notation, a compact format that preserves full semantic meaning while eliminating redundancy, cutting token usage by 30–65% with no workflow changes. For teams running thousands of AI operations daily, Toonify pays for itself immediately and keeps margins healthy as the product scales.

github.com/PCIRCLE-AI/toonify-mcp

WardnMesh

AI Security

As AI coding tools grow more autonomous, the question isn't whether they'll attempt dangerous operations — it's whether you'll know when they do. WardnMesh wraps any AI tool with a real-time security interception layer: 243 threat detection rules scan every input and output in under 5ms, pausing execution and asking for human confirmation before any high-risk operation proceeds. Fail-closed by design and fully local — no data ever leaves your machine — WardnMesh is the security foundation that lets teams embrace agentic development without surrendering control.

github.com/PCIRCLE-AI/wardnmesh

gstack-industrial

Developer Tooling

Claude Code skills are only valuable if you remember to use them. gstack-industrial solves the discovery problem with a smart router that analyzes your message and project state, then proactively suggests the right skill at the right moment — with anti-annoyance guardrails that keep it useful, not intrusive. A shared template system lets you update standards once and propagate changes across your entire skill library. As AI coding workflows mature into institutional assets with hundreds of skills, gstack-industrial is the intelligence layer that makes them usable day-to-day.

github.com/kevintseng/gstack-industrial

Yibook 易書

Divination Research · AI

Ancient divination systems — I Ching, BaZi, Zi Wei Dou Shu, Western astrology — encode millennia of pattern recognition, yet remain siloed and opaque. Yibook runs all four systems against a single query simultaneously, cross-references their conclusions, and surfaces convergence with transparent evidence chains — turning subjective interpretation into reproducible, multi-system analysis. As the global divination and spiritual wellness market surpasses $2B, Yibook positions itself as the first research-grade platform that treats these traditions as complementary data sources rather than competing belief systems.

yibook.pcircle.ai

Codex Factory Kit

Agentic Workflow

Most AI coding setups fail the same way: every session reconstructs the whole task from scratch, losing the decisions and context that make complex work coherent. Codex Factory Kit gives Codex a staged operating model — bootstrap, plan, implement, review, verify, document, retro — backed by durable in-repo artifacts (PRODUCT.md, PLAN.md, REVIEW.jsonl) that persist across sessions. The result is multi-session continuity, cleaner agent handoffs, and explicit review evidence: turning one-shot AI prompts into a compounding engineering workflow.

github.com/kevintseng/codex-factory-kit