Education & Research Skills

Generates unique algorithmic art using p5.js, emphasizing computational beauty and interactive exploration of generative aesthetics.

Generates thorough, cited research reports from multiple web sources using firecrawl and exa tools for in-depth analysis.

Equips users with advanced techniques for network protocol reverse engineering, including packet analysis and protocol documentation.

Conducts comprehensive research on GitHub repositories, generating structured markdown reports with timelines and metrics analysis.

Provides a systematic methodology for thorough web research, enhancing content generation with comprehensive, multi-angle insights.

Assists in writing publication-ready ML/AI papers with tools for citation verification and LaTeX templates for top conferences.

Evaluates LLMs using 60+ benchmarks for model quality assessment, widely adopted in academic and industry settings.

Enables researchers to interpret and manipulate PyTorch model internals, facilitating remote execution for massive models without local GPU resources.

Guides users in training and analyzing Sparse Autoencoders to uncover interpretable features in neural network activations.

Guides mechanistic interpretability research using TransformerLens to analyze transformer models and their internal mechanisms.

Enables zero-shot image segmentation using Meta AI's Segment Anything Model, allowing flexible object detection in various domains.

Guides LLM post-training with RL using slime, integrating Megatron-LM for custom data workflows and model training.

Guides users in implementing PyTorch-native reinforcement learning with torchforge, enabling scalable training and easy experimentation.

Provides a framework for generating high-quality sentence and text embeddings, supporting multilingual and multimodal applications.

Provides rigorous academic research engineering with a focus on theoretical correctness and optimal implementation across technologies.

Facilitates mass spectrometry data processing and analysis, enabling users to import, filter, and compare spectral data efficiently.

Provides a minimalist implementation of GPT-2 for educational purposes, enabling users to learn and experiment with transformer models.

Provides a comprehensive toolkit for UX research and design, enabling persona generation, journey mapping, and usability testing.

Enables querying Google NotebookLM for citation-backed answers, enhancing research efficiency and documentation management.

Automates audio-to-text transcription, generating professional Markdown documentation and summaries for meetings and lectures.