Kalyan Tunga Ramesh@Kalyan__tr

A gas-optimized lottery engine using Binary Search (O(log N)) to achieve 95% gas savings. Powered by Chainlink VRF v2.5 and Automation, it features a Factory architecture for permissionless, scalable, and verifiably fair prize distributions.

An ERC-4626 vault for tokenized RWAs implementing asynchronous T+1/T+2 settlement. Features stable NAV accounting, virtual liabilities for redemption queues, and rigorous stateful fuzzing to ensure solvency during settlement delays.

PGAP solves how to safely allow AI agents to spend real money without trusting the AI. As autonomous agents begin purchasing APIs, data, and services, critical risks emerge: overspending, hallucinated recipients, bypassed safeguards. Existing systems rely on off-chain trust or agent self-restraint, which doesn't scale safely. PGAP's answer: AI proposes payments, smart contracts enforce rules on-chain. ARCHITECTURE Three layers with clear trust boundaries: 1. Gemini AI Agent (untrusted proposer) - reads policy, proposes PaymentIntents 2. Backend Executor (untrusted relayer) - submits transactions without modification 3. TreasuryWithPolicy Contract (trusted enforcer) - validates all invariants on-chain POLICY ENFORCEMENT Multi-layer protection: - Per-transaction spending limits - Daily spending caps with automatic resets - Cooldown periods between payments - Recipient allowlists - Replay protection via nonces - Emergency pause mechanism Even if AI hallucinates, backend is compromised, or requests are spammed, funds remain safe. Worst-case loss is bounded by policy configuration. DEMO Five scenarios prove defense-in-depth: 1. Valid payment execution 2. AI refusal of over-limit requests 3. On-chain cooldown enforcement 4. Unauthorized recipient blocking 5. Nonce replay attack prevention This demonstrates AI reasoning plus immutable on-chain enforcement. PRODUCTION READY The $1 demo limit is intentionally conservative for testnet visibility. Production deployments configure limits based on risk tolerance—$100 for small businesses, $10,000+ for enterprises. Only configuration values change; enforcement logic remains identical. Built on Arc with native USDC settlement. Verified transactions on Arc Sepolia testnet prove end-to-end functionality. PGAP is a configurable policy engine enabling safe autonomous agent spending in production environments.

In this hybrid hackathon powered by Arc, developers will build the next generation of agentic commerce systems using Arc with USDC. Explore a world where AI agents and web services can pay for APIs, data, compute, and content in dynamic, usage-based ways, while Arc provides stablecoin-native settlement with predictable USDC fees and deterministic, sub-second finality. This hackathon is supported by Google & Google DeepMind, with a dedicated challenge and $40,000 in Google Cloud Platform (GCP) credits awarded for the best use of Gemini models and Google AI Studio. 📅 January 9 – 24, 2026 • Jan 9 – 23 (Online Phase) – Collaborate and build online with developers and AI innovators from around the world. All projects must be submitted by the end of the online phase on January 23rd. • Jan 23 (On‑site Build Day – San Francisco, CA) – Selected participants will be invited to an exclusive in‑person session to refine their projects and connect directly with mentors. • Jan 24 (On‑site Demos & Awards – San Francisco, CA) – Live pitching sessions to a panel of judges and ecosystem partners, followed by the official winner announcement. Get guidance from Arc and Circle experts through on-site support. 📍 On-site Venue (Jan 23–24): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 🤝 Go solo or team up, it's your flow. 🏆Compete for $50,000 in USDC & GCP Credits. 📍 On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑🏻‍💻 Secure your spot now - sign up before the Kick-Off Stream!