Phoenix - Your AI AssistantFoundation Model Prompting — Maximise Model Performance
Delivery in
4 days
- Views 1
Amount of days required to complete work for this Offer as set by the freelancer.
Rating of the Offer as calculated from other buyers' reviews.
Average time for the freelancer to first reply on the workstream after purchase or contact on this Offer.
What you get with this Offer
I will optimise your AI prompts for GPT, Claude, Gemini, or your preferred foundation model to improve accuracy, consistency, reasoning, and output quality — without fine-tuning.
Using a systematic testing approach, I’ll experiment with system prompts, few-shot examples, reasoning strategies, output formats, and model parameters to identify what works best for your specific use case.
What’s Included
✓ System prompt optimisation
✓ Prompt variant testing
✓ Few-shot example selection
✓ Reasoning prompt optimisation
✓ Output format & structure testing
✓ Temperature & top-p optimisation
✓ Evaluation using your real test data
✓ Performance comparison between prompt versions
✓ Final optimised prompt set
✓ Clear recommendations and documentation
Instead of relying on guesswork, I’ll use structured experimentation and evaluation to find the prompting approach that delivers the best results for your model and task.
Give your AI workflow better prompts, more reliable outputs, and stronger performance — without the cost of fine-tuning.
Using a systematic testing approach, I’ll experiment with system prompts, few-shot examples, reasoning strategies, output formats, and model parameters to identify what works best for your specific use case.
What’s Included
✓ System prompt optimisation
✓ Prompt variant testing
✓ Few-shot example selection
✓ Reasoning prompt optimisation
✓ Output format & structure testing
✓ Temperature & top-p optimisation
✓ Evaluation using your real test data
✓ Performance comparison between prompt versions
✓ Final optimised prompt set
✓ Clear recommendations and documentation
Instead of relying on guesswork, I’ll use structured experimentation and evaluation to find the prompting approach that delivers the best results for your model and task.
Give your AI workflow better prompts, more reliable outputs, and stronger performance — without the cost of fine-tuning.
What the Freelancer needs to start the work
Please provide:
Your current prompt(s) you want optimised.
The AI model you’re using (GPT, Claude, Gemini, etc.).
Your specific task or desired outcome.
Sample inputs and expected outputs, if available.
Evaluation data or examples showing where the current prompt performs poorly.
Any required output format or constraints.
If you don’t have evaluation data, simply provide your current prompt and explain what you want it to achieve.