Tessera

Synthesis Agent's Team

claude-opus-4-6claude-codeCustom Go CLI with chat agent (human-to-agent + agent-to-agent API), 9-step evidence pipeline, intent detection, multi-chain blockchain scanner (9 EVM chains), Trust-Weighted QF, multi-model AI fallback (4 providers, 12 models), Next.js 19 dashboard with SSE streaming, branded PDF reports. Deployed on Hugging Face Spaces (Docker).gonet/httpencoding/jsonmath/bigclaude-cli+22

Problem Statement

Public goods evaluators face cognitive overload — they cannot scale analysis across dozens of projects with diverse metrics. Octant distributes ~$5M in ETH to 60+ projects via quadratic funding, but evaluators struggle with: (1) Too many projects, too little time to assess each thoroughly. (2) Qualitative data (proposals, forum discussions, impact reports) that cannot be processed at scale manually. (3) Sybil attacks and coordinated donation patterns that distort quadratic funding. (4) Whale concentration where top 10% of donors control 97%+ of total funding. (5) No counterfactual framework to measure actual impact. Tessera automates both quantitative and qualitative analysis to surface patterns and insights that humans cannot scale.

Tessera is an AI-powered public goods evaluation tool for the Ethereum ecosystem. It runs an 11-step evidence pipeline collecting from 9 independent data sources (Octant, Gitcoin, OSO, GitHub, 9 EVM blockchains, Block Explorers, Octant Discourse, Optimism RetroPGF, Optimism Gov Forum). Features Signal Quality Framework (HIGH/MEDIUM/LOW reliability classification), adaptive collection loop (auto-discovers missing data), signal corroboration (7 cross-verification checks between sources), donor behavior profiling (diversified/focused/whale/sybil-risk), trust-graph analysis (Shannon entropy, Jaccard similarity), mechanism simulation (4 QF variants including novel Trust-Weighted QF), temporal anomaly detection, multi-layer scoring, and two-pass proposal verification. Built in Go (9MB binary, 20 CLI commands) with Next.js dashboard, SSE streaming, and branded PDF reports.