Orian B.

Key Skills

Software

Don't disclose

Other

Top Subject Matter

Accounting & finance auditing

AI evaluation for financial accuracy

LLM evaluation and RLHF for accounting/finance

Top Data Types

Text

Document

Top Task Types

RLHF

Freelancer Overview

Financial Analysis & AI Evaluation Specialist | Accenture (Finance & Technology Solutions). Brings 9+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Don't disclose. Education includes Bachelor of Business Administration, Texas A&M University – Mays Business School (2020) and Bachelor of Science, Northern Arizona University (2016). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

AI Evaluation & Data Science Specialist | Freelance (Multiple Clients incl. Deloitte AI Lab)

Don't discloseTextRLHF

Participated in RLHF workflows for LLMs by evaluating model financial responses, ranking candidate outputs, and providing detailed preference feedback to improve model alignment. Assessed hallucination risk and measured model performance using accuracy, precision, recall, and hallucination-rate metrics for financial AI tasks. Engineered and tested prompts to evaluate reasoning quality on accounting and analytical workloads. • Performed model response evaluation and preference ranking • Generated structured feedback for RLHF improvement • Built evaluation frameworks for finance-domain metrics • Tested prompts for reasoning alignment to accuracy requirements

2024 - Present

Financial Analysis & AI Evaluation Specialist | Accenture (Finance & Technology Solutions)

Text

Audited AI-generated financial summaries and statements by checking numerical accuracy, logical consistency, and correct interpretation against accounting expectations. Reviewed and verified expense classifications and financial dataset labels, identifying misclassifications and tagging discrepancies to ensure correct financial representations. Validated AI financial reasoning workflows (including tax logic and accounting rules) before approved training data and outputs were finalized. • Scored and compared model outputs for factual and policy alignment • Documented written explanations of corrections for traceability • Flagged analytical errors such as double-counted entries • Conducted QA-style verification to reduce recurrence of labeling mistakes

2024 - Present

AI Evaluation & Data Science Specialist | Freelance (Multiple Clients incl. Deloitte AI Lab)

OtherText

Reviewed and rated AI-generated text and structured financial outputs for factual accuracy, logical coherence, and policy compliance across varied project types. Established model evaluation frameworks to measure quantitative performance and detect hallucination patterns in financial AI systems. Conducted prompt-based model reasoning evaluation to support responsible model behavior in finance and analytics. • Rated outputs against accuracy and coherence criteria • Tracked metrics such as precision/recall and hallucination rates • Provided reasoning-level checks on finance statements • Supported iterative model improvement through evaluation findings

2022 - Present

Education

Work History