Finetuning-Forge
Local-first LLM fine-tuning workstation — Unsloth, Axolotl, and Torchtune backends with real-time GPU telemetry, live loss curves, and VRAM pre-flight checks.
LLMFine-tuningGPUNext.jsFastAPI
Client Demos
Live demonstrations of AI-powered products — from agentic pipelines to interactive visualizers.
EvoWrite
Agentic content generation platform with RAG, pgvector search, multi-LLM orchestration, and streaming pipelines.
RAGpgvectorLLMStreamingAgentic
Multi-Agent Research System
LangGraph-powered MARS that orchestrates five specialized AI agents (Planner, Researcher, Writer, Critic, Word) using Google Gemini to produce polished research reports.
LangGraphMulti-AgentFastAPIGeminiPython
Neural Visualizer
Interactive web app to build ANN, CNN, RNN, LSTM, GAN, Transformer, and Diffusion models and visualize architecture, propagation, and loss landscapes in real time.
Neural NetworksVisualizationPyTorchDashPlotly
Claw Review
AI-powered, read-only code review agent. Reads code, runs analyzers, emits structured findings, never mutates your source. Ships as a CLI + GitHub Action.
CLICode ReviewGitHub ActionsGroqPython