Open-Sourced Hackathon Search Agent Seeks Advice on Evaluation Architecture and Open-Source Models
AggressiveAerie9031 · reddit · 2026-08-12
A developer, tired of manually finding and applying to hackathons, open-sourced hackathon-searcher, a tool that discovers hackathons, researches them, scores them based on preferences, generates application answers, and handles the application flow. The focus now is the evaluation layer: combining structured info (location, travel support, prizes, themes, eligibility, deadlines) with user preferences to judge worth. He's debating between a single strong open-source LLM, a smaller local model with deterministic scoring, per-criteria scoring with separate aggregation, or multiple models; and which open-source models are good enough for structured judgment without slowing things down. He invites feedback from agent/LLM evaluation builders.
More from coding & agent
- Survey on Co-Evolution in Agentic Systems: Breaking Static Constraints — KyeGomezB · 2026-08-12
- OASIS: Automated Code Security Auditing Tool Powered by Ollama — tom_doerr · 2026-08-12
- W&B Shares Workflow to Block Prompt Injections and Alert Security Teams — wandb · 2026-08-12
- Conceptualizing 'SPVs' for Funding Single AI Agent Threads via Crypto — curious_vii · 2026-08-12
- Open-Source MCP Server: Free Web & Image Search Without API Keys — GoldBroccoli7073 · 2026-08-12
- Counter-intuitive AI Agent Behavior: Stop Extrapolating Human Thinking onto Agents — krishnan · 2026-08-12