Victor Krim

Key Skills

Software

AWS SageMaker

Anno-Mage

Appen

Argilla

Axiom AI

Clickworker

CloudFactory

Data Annotation Tech

Top Subject Matter

Coding problems.

Logic and mathematical reasoning.

Factual question answering.

Top Data Types

Computer Code Programming

Geospatial Tiled Imagery

Text

Top Task Types

Computer Programming Coding

Evaluation Rating

RLHF

Text Generation

Freelancer Overview

I'm an expert at all kinds of LLM review, especially coding and mathematical reasoning problems. My strengths are my excellent memory and attention to detail, making me well suited to fact-checking and consistency evaluation. I make sure to write descriptive and comprehensive reviews of model performance. I'm also skilled at writing prompts to test the boundaries of model harmlessness and safety restrictions.

Labeling Experience

Beryllium Multi-step Math and Reasoning

Data Annotation TechTextPrompt Response Writing SFT

The project involved guiding the LLM through a given multi-step Math or Reasoning problem. The goal was to modify the prompt to get the model as close as possible to a correct solution, and rewrite the model's Latex formatting to ensure it displays correctly.

The project involved guiding the LLM through a given multi-step Math or Reasoning problem. The goal was to modify the prompt to get the model as close as possible to a correct solution, and rewrite the model's Latex formatting to ensure it displays correctly.

2024

Sia Coding Prompts

Data Annotation TechComputer Code ProgrammingPrompt Response Writing SFT

Tasks consisted of writing coding prompts according to a specified category, difficulty, and complexity, and reviewing or correcting model responses.

Tasks consisted of writing coding prompts according to a specified category, difficulty, and complexity, and reviewing or correcting model responses.

2024 - 2024

Hera Factuality Evaluations

Data Annotation TechTextQuestion AnsweringText Generation

The project involved generating prompts or using pre-generated prompts, followed by A/B comparison of model outputs for Instruction Following, Truthfulness, Verbosity, Safety, and Overall Quality.

The project involved generating prompts or using pre-generated prompts, followed by A/B comparison of model outputs for Instruction Following, Truthfulness, Verbosity, Safety, and Overall Quality.

2023 - 2024

Achilles Data Analysis and Visualization

Data Annotation TechComputer Code ProgrammingQuestion AnsweringText Generation

The project involved asking the model a question about a csv dataset. The model responds by writing a Python/Pandas script, examining the output, and coming up with an answer to the question. Tasks consisted of review and scoring of model performance, and rewriting model answers.

The project involved asking the model a question about a csv dataset. The model responds by writing a Python/Pandas script, examining the output, and coming up with an answer to the question. Tasks consisted of review and scoring of model performance, and rewriting model answers.

2023 - 2023

Untitled labeling experience

No description provided.

No description provided.

Not specified

Education

Work History