#python

Miles is a fast, enterprise-ready tool for training large AI models after pretraining, with strong support for huge models, many GPUs, and both NVIDIA and AMD hardware. It helps you run training and rollout smoothly, update weights quickly, improve stability, and recover from failures without stopping the job. This can save time, reduce crashes, and make large AI training easier to manage.

https://github.com/radixark/miles

#python

This repo is a preserved archive of proof-of-concept security research and vulnerability writeups, with many folders copied from older repos and some added directly. The main benefit is that you can study real examples, compare fixes, and learn fuzzing and exploit research methods more easily. It also says the contents were checked for exact file matches, so the preserved PoCs should be reliable for review.

https://github.com/bikini/exploitarium

#python #ai #blackrock #finance #goldmansachs #hedgefund #jpmorgan #ml #quant #quantitative_finance #quantitative_trading #trading

AutoHedge is a tool that uses AI agents to trade for you, do market analysis, manage risk, and place orders with little human work. It now fully supports Solana and will add Coinbase and more exchanges soon. The benefit for you is faster trading decisions, built-in risk control, and a system that can scale to more markets and strategies.

https://github.com/The-Swarm-Corporation/AutoHedge

#typescript #ai_for_science #ai_research #ai_workbench #bioinformatics #claude_science_alternative #desktop_app #electron #linux #local_first #macos #mcp #model_agnostic #open_science #python #r #reproducible_research #research_software #research_workbench #scientific_ai #windows

AIPOCH Open Science is a local-first, open-source AI workbench for research on macOS, Windows, and Linux. It helps you plan, run, and track research in one place with files, code, data tools, reports, and clear proof of how results were made. The benefit to you is simpler, safer, and more repeatable research with less lost work and easier checking of results.

https://github.com/aipoch/open-science

#python #investment #quant #quantitative #quantitative_finance #trading

FinHack is a flexible tool for quantitative finance that helps with data collection, factor research, machine learning, strategy building, backtesting, and live trading. It is useful because it can save time, support different markets, and let you test ideas faster with tools for A-shares, U.S. stocks, futures, and crypto, plus multi-process speed and custom rules.

https://github.com/FinHackCN/finhack

#python #agent #llm #mathmodel #skills

This tool helps you do math modeling faster by finding, coding, fixing errors, and writing a full paper for you. You can use a desktop app, Docker, or local setup, and it works with many AI models. The benefit is clear: it saves time, gives ready-to-submit paper templates, and helps you produce cleaner, more complete work with less manual effort.

https://github.com/jihe520/MathModelAgent

#python #agents #agentskills #claude_code #deep_research #deep_research_agent

Hyperresearch is a tool that turns Claude Code into a deep research agent. It can search many sources, check citations, keep a reusable research vault, resume crashed runs, and help you create stronger reports with less manual work.

https://github.com/jordan-gibbs/hyperresearch

#python #ai #audio_generation #deep_learning #foundation_models #gpt #huggingface #llama #llms #music_generation #style_transfers #voice_cloning

YuE2 is a music generator that turns lyrics and a style prompt into an editable score, then into a full song with vocals and accompaniment. It can also make covers and let you change melody, harmony, tempo, or form before creating a new recording. This helps you create songs faster, inspect the music before final output, and control the result more easily.

https://github.com/multimodal-art-projection/YuE

#python #claude_ai #claude_pt #claude_skills #redteam #redteam_tools #skills

This is a library of ready-made security skill files that teach Claude offensive security tasks like web testing, identity abuse, wireless attacks, cloud checks, and red team work. You can add only the skills you need, so Claude stays focused and uses less context. For you, that means faster help for authorized security testing, bug bounties, training, and research, with clearer guidance and better results.

https://github.com/SnailSploit/Claude-Red