Lagus S.

Key Skills

Software

No software listed

Top Subject Matter

AI/LLM response evaluation and prompt quality assessment

AI training data labeling for LLM pipelines (instruction following, response ranking, quality assessment)

Top Data Types

Text

Document

Top Task Types

RLHF

Entity (NER) Classification

Freelancer Overview

AI Data Annotator & Prompt Evaluator. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. AI-training focus includes data types such as Text and Document and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

AI Data Annotator & Prompt Evaluator at Outlier AI (2022–Present)

TextRLHFEntity (NER) Classification

Evaluated and ranked AI-generated responses for accuracy, coherence, instruction-following, and overall quality across multiple task categories. Applied rigorous quality checks to maintain accuracy rates consistently above platform benchmarks while handling complex multi-step prompts and responses. Identified edge cases and ambiguous outputs and provided structured feedback to improve annotation and evaluation guidelines. • Assessed writing, reasoning, and coding-related response quality • Performed factual research, mathematical reasoning, and creative writing evaluation • Flagged edge cases and ambiguous outputs for guideline refinement • Used Python-based scripts and internal tooling to manage annotation batches

2022 - Present

Data Labeling Specialist at Handshake (AI Tasks) (2021–Present)

DocumentRLHFEntity (NER) Classification

Completed high-volume labeling and evaluation tasks for AI training pipelines, including instruction-following evaluation, response ranking, and content quality assessment. Applied consistent annotation standards across diverse document types to meet strict accuracy and formatting requirements. Collaborated in structured tasking environments to deliver reliable labeled datasets under tight deadlines while adapting to evolving guidelines. • Ranked responses and assessed instruction-following and content quality • Maintained strict accuracy and formatting requirements for labeled outputs • Delivered scalable labeled datasets in deadline-driven, tasking workflows • Adapted quickly to changing task guidelines and project priorities

2021 - Present

Work History