# Degen Den AI Research Platform
> https://ai.degenden.xyz

Browser-native AI research workstation. 37 interactive labs across 7 tracks.
100% client-side compute. Zero uploads. Polygon-verified. Free to use.

## Platform
- App: https://ai.degenden.xyz
- 37 labs, 7 tracks
- All computation runs in the browser

## Research Tracks

### Structured Data Science Labs
> https://ai.degenden.xyz/structured-data/
Browser-native CSV and Excel analytics: upload, profile, clean, visualize, and ask natural language questions about your data — all on-device, zero server.

- [CSV Data Science](https://ai.degenden.xyz/modules/csv-ds/) — Upload any CSV and run full exploratory data analysis in your browser: column profiling, n
- [Excel Data Science](https://ai.degenden.xyz/modules/excel-ds/) — Multi-sheet Excel analytics engine running 100% in-browser: pivot table builder, time seri

### Machine Learning Labs
> https://ai.degenden.xyz/machine-learning/
Interactive ML algorithm simulations: regression, classification, clustering, feature engineering, and model evaluation — all running in your browser with real data.

- [Feature Engineering](https://ai.degenden.xyz/modules/ml-feature/) — Interactive feature engineering pipeline: column profiling, one-hot and label encoding, mi
- [Regression](https://ai.degenden.xyz/modules/ml-regression/) — Simulate OLS, multiple linear regression, polynomial regression degrees 1-3, Ridge and Las
- [Classification](https://ai.degenden.xyz/modules/ml-classify/) — Run and compare kNN, Logistic Regression, Naive Bayes, and Decision Tree classifiers with 
- [Clustering](https://ai.degenden.xyz/modules/ml-cluster/) — Interactive clustering workstation: k-Means++ with elbow method, DBSCAN with epsilon and m
- [Model Evaluation](https://ai.degenden.xyz/modules/ml-evaluate/) — Complete model evaluation toolkit: bias-variance decomposition with bootstrap, learning cu

### Deep Learning Labs
> https://ai.degenden.xyz/deep-learning/
Visualize neural network architectures, backpropagation, CNNs, LSTMs, attention mechanisms, and training dynamics — step by step, parameter by parameter.

- [Neural Networks](https://ai.degenden.xyz/modules/dl-neural/) — Build and compare neural network architectures: choose layers, activations, and optimizers
- [Activations](https://ai.degenden.xyz/modules/dl-activations/) — Compare ReLU, Sigmoid, Tanh, Swish, GELU, and Leaky ReLU: visualize saturation zones, deri
- [Backpropagation](https://ai.degenden.xyz/modules/dl-backprop/) — Step through the chain rule computation, visualize vanishing and exploding gradients, comp
- [Regularization](https://ai.degenden.xyz/modules/dl-regularize/) — Live regularization experiments: Dropout rate effects, Batch Normalization behavior, L1/L2
- [CNNs](https://ai.degenden.xyz/modules/dl-cnn/) — Interactive CNN visualizer: apply convolution kernels to images, watch feature maps emerge
- [RNNs & LSTM](https://ai.degenden.xyz/modules/dl-rnn/) — Simulate RNN, LSTM, and GRU cell operations step-by-step: visualize forget, input, and out
- [Attention](https://ai.degenden.xyz/modules/dl-attention/) — Interactive attention mechanism explorer: compute self-attention QKV matrices step-by-step
- [Training Dynamics](https://ai.degenden.xyz/modules/dl-training/) — Compare optimizer trajectories on loss landscapes, experiment with learning rate schedules

### Unstructured Data Labs
> https://ai.degenden.xyz/unstructured-data/
Analyze documents, images, audio, and video in-browser: NLP pipelines, FFT spectrograms, pixel inspection, scene detection — zero server uploads.

- [Document Data](https://ai.degenden.xyz/modules/document/) — Full document NLP pipeline in-browser: readability scores, TF-IDF keywords, per-sentence s
- [Image Data](https://ai.degenden.xyz/modules/image/) — Explore images at the pixel level: 17 real-time canvas filters, R/G/B channel isolation, i
- [Audio Data](https://ai.degenden.xyz/modules/audio/) — Browser-native audio analysis: waveform viewer, Cooley-Tukey FFT spectrum, STFT spectrogra
- [Video Data](https://ai.degenden.xyz/modules/video/) — In-browser video analytics: scene detection by frame difference, motion intensity via opti

### Generative AI Labs
> https://ai.degenden.xyz/generative-ai/
Simulate GANs, VAEs, diffusion models, LLM sampling, fine-tuning techniques, and evaluation metrics for generative AI — all in one interactive workstation.

- [GANs](https://ai.degenden.xyz/modules/ge-gan/) — Simulate GAN training dynamics: generator vs discriminator loss curves, mode collapse visu
- [VAEs](https://ai.degenden.xyz/modules/ge-vae/) — Explore variational autoencoder mechanics: ELBO loss decomposition, KL divergence visualiz
- [LLM Lab](https://ai.degenden.xyz/modules/ge-llm/) — Experiment with LLM internals: tokenization explorer, temperature and top-p sampling visua
- [Diffusion](https://ai.degenden.xyz/modules/ge-diffusion/) — Simulate DDPM forward and reverse diffusion: noise schedule comparison, U-Net denoising st
- [Evaluation](https://ai.degenden.xyz/modules/ge-eval/) — Comprehensive generative AI evaluation: FID scores for images, BLEU/ROUGE for text, BERTSc
- [Fine-tuning](https://ai.degenden.xyz/modules/ge-finetuning/) — Explore LLM fine-tuning techniques: LoRA rank and alpha configuration, QLoRA 4-bit quantiz

### Agentic AI Labs
> https://ai.degenden.xyz/agentic-ai/
Explore autonomous AI agents: ReAct loops, memory systems, tool use, multi-agent orchestration, and full RAG pipelines — the architecture behind modern AI assistants.

- [Agent Core](https://ai.degenden.xyz/modules/ag-core/) — Simulate the complete agent loop: perceive-think-act cycles, ReAct reasoning traces, agent
- [Agent Memory](https://ai.degenden.xyz/modules/ag-memory/) — Explore all agent memory types: in-context short-term, external long-term, episodic buffer
- [Tool Use](https://ai.degenden.xyz/modules/ag-tools/) — Build and simulate function calling schemas, parallel tool execution graphs, error handlin
- [AI Planning](https://ai.degenden.xyz/modules/ag-planning/) — Chain-of-Thought step generation, Tree of Thought branching with evaluation, Monte Carlo T
- [Multi-Agent](https://ai.degenden.xyz/modules/ag-multiagent/) — Visualize multi-agent architectures: orchestrator-specialist networks, agent debate simula
- [RAG](https://ai.degenden.xyz/modules/ag-rag/) — Full RAG pipeline simulator: chunking strategy comparison, vector embedding visualization,

### Quantum AI Labs
> https://ai.degenden.xyz/quantum-ai/
Interactive quantum computing from first principles: Bloch sphere, quantum circuits, QML, QAOA optimization, hybrid classical-quantum AI, and fault tolerance roadmap.

- [Qubit](https://ai.degenden.xyz/modules/q-qubit/) — Interactive Bloch sphere with 3D isometric projection, qubit state controls, quantum gate 
- [Quantum Circuits](https://ai.degenden.xyz/modules/q-circuits/) — Build quantum circuits with exact 2-qubit state vector simulation: preset Bell states, GHZ
- [Quantum ML](https://ai.degenden.xyz/modules/q-qml/) — Variational quantum classifier with param-shift gradient training, quantum kernel vs class
- [Quantum Optimizer](https://ai.degenden.xyz/modules/q-optimizer/) — QAOA MaxCut simulation with graph visualization, simulated quantum annealing, 5-city TSP t
- [Hybrid AI](https://ai.degenden.xyz/modules/q-hybrid/) — Animated classical-quantum variational loop, VQE H2 molecule ground state simulation, zero
- [Quantum Future](https://ai.degenden.xyz/modules/q-future/) — Interactive quantum timeline from 1994 to 2042, surface code grid simulation, QEC code com

## Technical
- React SPA, browser-native compute
- Quantum: exact 2-qubit state vector in JS
- ML: pure JS — no Python backend
- Blockchain: Polygon network

## About
Creator: Degen Den XYZ — https://degenden.xyz
Sitemap: https://ai.degenden.xyz/sitemap.xml