AunaliakaCossale

Within 2 months into joining InvestBrains, I have been involved in a variety of projects, including:

• Fine-tuned Large Language Models (LLMs) for specialized tasks including role-play scenarios and question-answering systems. This work improved model performance for specific use cases.

• Worked on fine-tuning diffusion models, particularly Stable Diffusion, for custom image generation tasks. This involved creating and curating datasets tailored to our project requirements.

• Implemented Direct Preference Optimization (DPO) techniques to better align models with user preferences, resulting in improved output quality and user satisfaction.

• Developed and deployed a moderation API utilizing vision models to detect harmful content, addressing important safety concerns in AI applications.

• Created and maintained automation scripts for LLM hosting, benchmarking, and synthetic data generation. These efforts streamlined our development processes and improved team efficiency.

• Engaged in ongoing learning and application of emerging AI techniques, staying current with advancements in the field.

• Website

• Python
• PyTorch
• HF Transformers/Diffusers/Datasets
• Axolotl
• Unsloth
• TRL

• Worked with a globally distributed team of 15 developers to build ReVanced, an open-source Android modification framework empowering users to customize apps to suit their needs.

• I helped spearhead overall architecture, design, and roadmapping - devising an adaptable core to enable endless customizations for apps built on Dalvik like YouTube. I also directly contributed extensive code across the codebase.

• I led development of the ReVanced Manager app installed by over 170K users. The app streamlined installations and made core functionality easily accessible.

• Initially adopted by a few hundred users, ReVanced gathered incredible traction on platforms like Reddit, Discord and Telegram from folks who really liked what we built.

• As installs grew exponentially to 170,000+, I took on expanded duties - overseeing roadmap priorities, debugging complex issues, liaising with user communities, and ensuring stability through rigorous testing.

• The project gave me hands-on experience on how coordinated remote teams can build delightful products loved by users globally. Debugging performance problems taught me how to approach issues methodically. Ultimately, the ability to solve real user problems at scale was extremely fulfilling.

• Website GitHub

• Flutter
• Dart
• Kotlin
• JADX