Brofara 1.

Key Skills

Software

No software listed

Top Subject Matter

LLM output evaluation

prompt engineering quality assurance

Prompt engineering and LLM workflow logic design

Top Data Types

Text

Top Task Types

Prompt + Response Writing (SFT)

RLHF

Freelancer Overview

AI response evaluation and grading using rigid logical workflow. Core strengths include Internal, Proprietary Tooling, and Google Workspace (Docs. AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

Systematic logic & parameters testing (robustness evaluation)

TextRLHF

Built robustness test frameworks to stress continuous logic models over large iteration counts, validating consistency and anomaly handling. Defined strict constraint parameters for AI engines to analyze anomalies and maintain data consistency. Used massive cycle testing (up to 20,000 iterations) to improve output accuracy under continuous evaluation. • Robustness evaluation of logic models across many iterations • Constraint parameter design for anomaly detection • Data consistency checks to improve accuracy • Iterative testing to validate logical integrity

Present

AI-assisted automation workflow design (logic formulation & prompt engineering)

TextPrompt + Response Writing (SFT)

Designed operational prompts and logic workflows to guide LLM behavior in generating scripts integrated with external systems. Used iterative prompting techniques to translate complex operational rules into structured instructions for models. Verified and tested logic flows to ensure accurate, lossless human-AI collaboration within defined constraints. • Prompt engineering for structured, rule-heavy instructions • Iterative prompt refinement to produce operational scripts • Integration-oriented logic formulation with Google Sheets API • Verification/testing to ensure zero data loss and correct flow

Present

AI response evaluation and grading using rigid logical workflow

Text

Evaluated and graded AI model outputs by applying strict, multi-step logic and quality-control checks. Identified factual errors, anomalies, formatting issues, and subtle inaccuracies, then provided clear logical feedback to improve response quality. Focused on context verification and bias/error detection to support reliable model behavior. • Quality control and fact-checking of AI-generated responses • Context verification and anomaly spotting in outputs • Error and bias detection with structured reasoning • Feedback and grading to determine superior answers

Present

Education

Work History