Browse categories
Explore
Fiverr Pro
English
$
USD
I will do ai workflow evaluation and llm prompt optimization with chatgpt and claude
Davex Scott
About this gig
GETTING INCONSISTENT RESULTS FROM CHATGPT, CLAUDE OR ANOTHER LLMS?
I provide AI workflow evaluation and prompt optimization to identify failure points, test outputs, and improve the reliability of your AI process.
I don't just rewrite prompts. I examine how your prompts, inputs, workflow logic, and AI outputs work together, then use representative tests to identify what needs improvement.
WHAT I CAN EVALUATE
- AI prompts and prompt systems
- ChatGPT and Claude workflows
- LLM output quality and consistency
- Multi-step AI workflows
- AI agent workflows
- Internal AI processes
DEPENDING ON YOUR PACKAGE, YOU CAN RECEIVE:
- AI workflow or prompt audit
- Evaluation criteria and testing
- Failure-point analysis
- Prompt optimization
- Before/after output comparisons
- Prompt library
- Workflow documentation
- Handoff guidance
Ideal for businesses, agencies, founders, teams, and professionals already using AI who want more consistent and testable results.
Send me your current prompt, workflow, sample outputs, or use case and I'll recommend the right scope.
Get to know Davex Scott
Davex Scott
AI Automation Expert for AI Agents and Video Workflows That Save Business Time
- FromUnited Kingdom
- Member sinceJul 2026
- Avg. response time22 hours
Languages
English, French, Spanish
AI AUTOMATION for practical business workflows, AI agents, lead qualification, and content processes.
I build AI-powered workflows using ChatGPT, Claude, and CapCut, with a focus on AI agent development, process automation, workflow design, AI video creation, and video editing.
I can help turn repetitive processes and content requirements into structured, practical AI workflows.
Turnaround: 3 days | Revisions: 1
Tell me what workflow or content process you want to improve, and let’s build a practical solution.
My Portfolio
Other AI Development Services I Offer
FAQ
What is included in your AI workflow evaluation?
I review your prompts, workflow logic, inputs, outputs, and known failure cases within the package scope. I then identify inconsistencies and recommend or implement improvements.
Can you evaluate ChatGPT and Claude workflows?
Yes. I can evaluate existing ChatGPT, Claude, and other LLM workflows based on your inputs, desired outputs, and evaluation criteria.
Do you only rewrite prompts?
No. The focus is evaluation and improvement. I examine how the prompt performs within the workflow and use representative outputs to identify weaknesses.
Can you evaluate an AI agent workflow?
Yes. I can review AI agent workflows and evaluate prompt behavior, outputs, workflow steps, and recurring failure points within the agreed scope.
How do you determine whether a prompt improved?
I compare representative outputs before and after optimization using defined evaluation criteria where appropriate.
Can you create a prompt library?
Yes. Standard and Premium can include a structured prompt library documenting the improved prompts and their intended use.
What do you need from me?
Ideally, your current prompts or workflow, representative inputs and outputs, your desired results, and any known problems or failure cases.
Which package should I choose?
Choose Basic for a focused audit. Choose Standard for testing and evaluation of a broader workflow. Choose Premium for comprehensive evaluation, optimization, testing, and documentation.
