Professional Overview
LLM Evaluation specialist — 5 years building evaluation harnesses, benchmark sets, and regression frameworks at Apple. I measure what others demo: accuracy, consistency, hallucination rates, edge-case handling. Built eval workflow tool, 25% error reduction and a 287-vector security benchmark. Python, RAG, prompt engineering, SQL.
Experience
AI Expertise
Skills
Employment Type
Description of Responsibilities
- Built repeatable evaluation workflows for LLM-powered voice assistant features
- Created benchmark datasets from real historical cases establishing ground-truth
- Designed failure analysis frameworks to uncover edge cases and bias patterns
- Partnered with product teams to translate business requirements into measurable
Key Achievements / KPIs
- Developed Assistant - reduced grader error rates by 25%
- Built security benchmark with 287 attack vectors for LLM system prompts
- Delivered quality evaluation across 5+ LLM model generations
AI Expertise
Skills
Employment Type
Description of Responsibilities
- Evaluated LLM outputs across multilingual voice assistant scenarios, identifying
- Built PODE Assistant — AI-assisted evaluation tool, reducing grader error rates
- Implemented LLM-as-Judge evaluation: designed judge prompts, validated alignment
- Curated golden datasets from production interactions, built regression test feed
- Compared model performance across OpenAI, Anthropic, on-device: cost, latency, q
Key Achievements / KPIs
- Reduced grader error rates by 25% with PODE Assistant toolContributed to evaluat
- Built System Prompt Benchmark: 287 attack vectors for LLM security evaluation
AI Expertise
Skills
Employment Type
Description of Responsibilities
- Designed evaluation frameworks and performance metrics for distributed teams
Education
Field of Study: AI Agents and Agentic AI with Python and Generative AI
Field of Study: Prompt Engineering for ChatGPT
Field of Study: Marketing
Areas of Expertise
Automation & AI Workflows
Build intelligent automation pipelines, integrate AI APIs into existing tools, and replace manual workflows with agentic AI systems.
NLP & Prompt Engineering
Build natural language processing systems, fine-tune large language models, and craft reliable production prompts for AI-powered products.
MLOps & AI Infrastructure
Architect scalable ML infrastructure — CI/CD for ML, model versioning, experiment tracking, monitoring, and cloud deployment on AWS, GCP, or Azure.
