Danilo Batson

About

AI/LLM engineer with crypto/social-data domain experience who built an end-to-end AI trading signals and analysis backend using Next.js, Inngest, Gemini, and Supabase, including real-time progress UX and robust rate-limit/reliability handling. Also published an MCP SDK on NPM for authenticated agent tool access and built autonomous Twitter-based AI agents; strengthened reliability at LunarCrush by adding CI-gated E2E testing across web and mobile (Cypress/Maestro).

Experience

Key Strengths

• Built and shipped production-critical AI/LLM trading signals pipeline end-to-end (prototype to deployment)
• Strong LLM reliability engineering (strict typed parsing, validation, safe fallbacks for malformed outputs)
• Designs around API rate limits with batching/delays and retryable step orchestration
• Delivers strong UX for AI workflows via real-time progress tracking (Supabase subscriptions)
• Owns ambiguous projects end-to-end (scoped and shipped an AI Twitter crypto-analysis bot with no spec/budget)
• Pragmatic tradeoff decision-making (latency vs cost; quality vs speed; concise outputs within constraints)
• Built agent/tooling infrastructure (MCP SDK on NPM enabling authenticated agent access to platform data)
• Improved production reliability by implementing CI-gated E2E testing across web/mobile, preventing regressions