Resolve

AI-assisted opportunity filter for freelancers: detect stale bounties, scam repos, claimed rewards, and low-value tickets before wasting time.

Comment

Inspiration

Resolve was built from a real operational problem: finding paid technical work is not just about spotting opportunities, it is about knowing which ones are actually worth pursuing.

While looking for small and medium paid tickets, we found many opportunities that looked attractive but were already claimed, closed, over-competitive, expired, region-restricted, or unsafe. During this Build Week, we also received a blockchain client ZIP that appeared to contain EVM and Solana contracts. A static inspection showed suspicious repository behavior, including a Git hook that attempted to download and execute external code. That turned Resolve into a safety-first developer tool.

What it does

Resolve helps freelancers and small technical teams decide whether to apply, investigate, or reject paid opportunities before spending hours on them.

It classifies opportunities as:

  • Apply now
  • Investigate
  • Strategic job
  • Reject

It checks for practical blockers such as:

  • closed issues
  • missing repositories
  • claimed or paid bounties
  • excessive submissions or competing pull requests
  • expired deadlines
  • region restrictions
  • required upfront capital
  • hourly value below the target floor
  • suspicious repository packaging

How we built it

Resolve is a lightweight Node.js CLI with structured JSON sources for marketplace scans. The first version evaluates real candidates from Algora, Superteam Earn, Telegram job channels, and manual intake.

Core components include:

  • a scoring and classification script
  • structured opportunity source files
  • rejection reasons and priority labels
  • a workflow for safe manual review before executing unknown code

How we used OpenAI and Codex

Codex helped us design the scoring model, implement the first working filter, inspect real bounty opportunities, validate stale GitHub and Algora issues, and safely analyze suspicious client files without running them.

This project is not just a tracker. It is an AI-assisted decision layer for turning public opportunities into safer, more focused freelance income.

Challenges

The hardest part was separating attractive-looking opportunities from truly payable work. A bounty can show a high price and still be a bad target if it has already been claimed, has too many competing pull requests, or requires work outside the safe scope.

What we learned

For freelance income, the first product is judgment. Resolve helps developers protect their time, protect their machines, and focus only on opportunities with a real chance of getting paid.

What's next

Next we want to add automated GitHub and Algora validation, browser-based intake, a web dashboard, daily scans for fresh bounties, and a safer sandbox workflow for reviewing client-provided code.

Built With

Share this project:

Updates