Description
Needs to hire 10 Freelancers Summary We are looking for working AI engineers and consultants to put a new model evaluation tool through real use, on your own work, and tell us honestly where it falls short. RedCrown.ai runs one task across multiple models and configs, scores the outputs against your ground truth on cost, quality, and latency, and produces a shareable proof page naming the cheapest config that clears your quality bar. WHAT YOU DO (about 45 minutes) 1. Set up an account and connect your own provider API key (OpenAI, Anthropic, or another supported provider) in the Connection Hub. This is what lets the evaluation run against the models and data you actually choose, and anyone who fits the profile below already has a key handy. 2. Run one evaluation on a real task from your own work. Not a demo, not a toy prompt. Something you have actually shipped or are deciding about right now. 3. Share the resulting proof link with us. 4. Fill out a short feedback form at redcrown.ai/beta WHAT YOU GET - $25 as a thank-you for your time, on completion. This is not meant as a wage, the real value is below. - One month of the Growth plan, free. Unlimited runs, 10 active experiments, and the live proxy unlocked. No card required, and nothing charges after it ends. - A proof page you can put in front of your own client. WE ARE LOOKING FOR PEOPLE WHO - Have shipped an LLM feature for a paying client. - Have an opinion about which model they default to, and why. TO APPLY Answer the two screening questions. Applications without real answers to both will not be considered.