AI Research and Prompt Optimization

Overview and Summary

Partnered with the Managing Director to evaluate and apply AI tools in a real consulting practice. Conducted hands-on research across leading LLMs, pressure-tested them against specific use cases, and translated findings into practical recommendations that improved efficiency and decision-making.

Key Features

LLM Evaluation and Comparison

Evaluated and compared leading LLMs (Claude, ChatGPT, Gemini, Perplexity, NotebookLM) across real-world criteria including prompt behavior, strengths/limitations, and workflow fit.

Prompt Refinement

Ran structured discovery to understand AI usage in practice, then refined prompts and workflows to improve output quality and reduce iteration time.

Data Validation and Research

Consolidated and validated AI-assisted outputs into clean, usable datasets for competitive and VC research.

Strategic Guidance

Synthesized findings into clear, use-case-driven guidance on when to use (and not use) specific models.

Consulting Integration

Translated research findings into practical recommendations that improved consulting practice efficiency and decision-making.

Technologies and Skills

AI Tools and Platforms

Claude, ChatGPT, Gemini, Perplexity, NotebookLM

Research and Analysis

AI Research, Model Comparison, Data Validation

Problem-Solving

Prompt Refinement, Structured Analysis, Use-Case Optimization

Communication

Technical Documentation, Clear Communication of Tradeoffs

Implementation Highlights

Comprehensive LLM Evaluation Framework

Designed and executed a systematic evaluation framework that tested multiple leading LLMs against real consulting use cases, documenting strengths, limitations, and ideal applications for each model.

Prompt Engineering and Workflow Optimization

Discovered how AI was being used in practice through structured interviews and testing, then refined prompts and workflows to significantly reduce iteration time and improve output quality.

Strategic Recommendations and Insights

Translated complex research findings into actionable, clear guidance for the Managing Director on model selection and implementation strategies tailored to specific consulting scenarios.

Learning Outcomes

Gained deep understanding of different LLM capabilities, limitations, and behavioral characteristics across multiple platforms
Developed expertise in prompt engineering and optimization techniques for improving AI output quality
Learned to conduct structured research and analysis to evaluate complex technology solutions
Improved ability to translate technical findings into clear, actionable business guidance
Enhanced problem-solving skills through pressure-testing AI tools against real-world consulting use cases
Practiced professional communication and documentation in a consulting environment

Back to Experience