Dermot H.

Key Skills

Software

Other

Top Subject Matter

AI agent evaluation (LLM reasoning and RAG)

Software Quality Assurance, Python, Java

Math - Vector Calculus, Linear Algebra, Discrete

Top Data Types

Document

Text

Image

Top Task Types

Text Generation

Fine-tuning

Classification

Computer Programming/Coding

Evaluation/Rating

Function Calling

RLHF

Freelancer Overview

Privacy-First Local AI Agent (GenAI & MLOps) — human + LLM-as-a-judge evaluation. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Python, Java, Software development, and advanced mathematics. Education includes Bachelor of Science, Concentration in AI, Purdue University Northwest (2026) and Associate in Applied Science, Purdue University Northwest (2026). AI-training focus includes data types such as Document and labeling workflows including Evaluation and Rating.

Labeling Experience

Privacy-First Local AI Agent (GenAI & MLOps) — human + LLM-as-a-judge evaluation

OtherDocument

Conducted human and LLM-as-a-judge evaluations to assess reasoning quality and retrieval-augmented generation outputs. Evaluated Chain-of-Thought (CoT) reasoning and RAG accuracy against predefined criteria and guardrails. Used these evaluations to identify failure modes and inform iterative improvements to the local agent workflow. • Assessed CoT reasoning quality using human review and automated judging • Checked RAG accuracy using retrieved context from a local vector store • Applied prompt and guardrail constraints to standardize evaluation • Reported and refined results through an iterative model engineering loop

2026 - 2026

Education

Work History