UNDER CONSTRUCTION

Site currently being updated...

Check back soon for the full experience

    ___________
   /          /|
  /          / |
 /__________/  |
 |   ____   |  /
 |  |    |  | /
 |__|____|__|/
            

gabriella@kryptive.com

GitHub Medium LinkedIn YouTube Ko-fi

RESUME

GABRIELLA BARIS

AI Safety Researcher | Tool Developer

gabriella@kryptive.com | gabriellabaris.com

SUMMARY

Transitioning from CGI/animation to AI safety research, leveraging creative problem-solving skills and technical expertise to contribute to responsible AI development through red teaming, safety evaluation, and tool development for the AI community.

TECHNICAL SKILLS

Research

Red teaming, safety evaluation, vulnerability assessment, adversarial testing

AI/ML

Fine-tuning (LoRA/QLoRA), Hugging Face Transformers, PyTorch, Unsloth

Programming

Python, API integration, Web Scraping, Data Processing, Tool Development, Pipeline Automation

AI RESEARCH & PROJECTS

Red Teaming Research

OpenAI GPT-OSS-20b Red-Teaming Challenge Participant

Kaggle Competition | 2024

• Discovered critical vulnerabilities including system prompt injection causing deceptive behavior in CoT reasoning

• Identified paradoxical prompt patterns triggering infinite loops in chain-of-thought processing

• Found evaluation awareness issues where models prioritize passing tests over factual accuracy

• Documented CBRNE jailbreak methods through creative writing prompts

• Developed automated testing framework for red team findings

AI Development Tools

LLM Scribe

Dataset Creation Tool | 2024

• Built professional toolkit for creating high-quality conversational datasets for LLM fine-tuning

• Implemented support for multiple export formats (ChatML, Alpaca, ShareGPT/Vicuna)

• Integrated real-time token tracking with popular tokenizers (OpenAI, HuggingFace, Mistral)

• Features auto-save functionality, progress tracking, and format-specific customizations

Discord Bot for Local LLM Systems

2024

• Developed integration for Ollama and LM Studio

• Enabled seamless interaction with local language models through Discord interface

Model Fine-Tuning & Dataset Development

• Fine-tune local models using LoRA and QLoRA techniques

• Create custom datasets for specialized model training

• Implement data collection and preprocessing pipelines

WORK EXPERIENCE

3D Artist

AD&M | July 2024 - Present | Freelance

• Modeled 3D assets and environments for marketing materials

• Developed environmental scenes and camera animations

Character Artist

Onchain Studios | January 2023 - November 2023 | Full Time

• Created character models and textures for digital collectibles

• Optimized characters for Unity implementation

Visualization Artist

Technicolor Preproduction / MPC | October 2021 - September 2022 | Contract

• Visualization artist on films including Nope, Transformers Rise of the Beasts, and Dune: Part Two

• Animated and lit characters and objects in Maya

3D Artist

Sea Ranch Labs | June 2021 - August 2021 | Freelance

• Modeled and generated 3660 NFT assets

• Built automated pipeline for asset generation

3D Artist

Wacky World Studios | July 2020 - November 2020 | Freelance

• Created character rigs and animations

• Implemented rendering pipelines

EDUCATION

Bachelor of Fine Arts - Computer Animation

Ringling College of Art & Design | Sarasota, FL | May 2020
Download PDF Version →