manan-tech/legal-document-assistant
Legal Document Assistant: RAG over the Code of Civil Procedure and the Army Act using LangChain
Curious about Machine Learning and Artificial Intelligence.
Legal Document Assistant: RAG over the Code of Civil Procedure and the Army Act using LangChain
A Python library for advanced quantitative portfolio analysis, optimization, and validation.
Local-first token-efficient routing agent — AMD Developer Hackathon ACT II Track 1
Multi-agent RL sim: 4 traders, market maker & SEC regulator trained via GRPO on Llama-3.2. 250 steps of emergent financial crisis — slaughter, adaptation, collusion, regulatory oversight — no scripts, just learned behavior.
AI Agent Framework, the Pydantic way
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Flutter makes it easy and fast to build beautiful apps for mobile and beyond
:house_with_garden: Open source home automation that puts local control and privacy first.
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
Production-ready platform for agentic workflow development.
Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]
Open source platform for AI Engineering: OpenTelemetry-native LLM Observability, GPU Monitoring, Guardrails, Evaluations, Prompt Management, Vault, Playground. 🚀💻 Integrates with 50+ LLM Providers, VectorDBs, Agent Frameworks and GPUs.
PyTorch WGAN implementation on CIFAR-10. Features a GPU-optimized training script with weight clipping and a premium Streamlit dashboard for interactive latent manifold interpolation, synthetic generation, and theory walkthroughs. Perfect for exploring Generative AI.
AI-powered investment research platform using multi-agent LLMs (LangGraph + AutoGen) to generate institutional-grade memos with risk scoring, network analysis & consensus voting.
SIH 2025 language agnostic chatbot sample
“A Streamlit app that combines Arxiv papers with LLMs using Retrieval-Augmented Generation (RAG). Search papers, get context-aware answers, and explore titles with expandable summaries. Powered by FAISS embeddings, HuggingFace models, and OpenRouter LLMs (Llama 4, Deepseek R1, GPT OSS).”
Trained a model using Bidirectional LSTM and trained it on Hamlet. Then created app with streamlit to predict next word of hamlet sentence.
This is to create a code that will automatically summarise any topic input by used. This uses llama3.2 model running locally on Ollama.
Kaggle | Playground S04E08
Kaggle Pneumonia detection using Tensorflow 2.0
Config files for my GitHub profile.