README
Solana Clawd NVIDIA Trading Factory Instruct
Specialized SFT data for a Solana-native NVIDIA algorithmic trading factory. It teaches data ingestion, GPU feature engineering, alpha research, cuML KDE scenario generation, cuFOLIO/cuOpt Mean-CVaR optimization, paper execution policy, risk controls, backtesting, monitoring, and Clawd governance.
Format
Each row uses OpenAI-style messages plus metadata:
{"messages": [{"role": "system", "content": "..."}, {"role": "user", "content": "..."}, {"role": "assistant", "content": "..."}], "metadata": {...}}
Splits
Produced by scripts/prepare_dataset.py with seed 42.
| Split | Examples |
|---|---|
| train | 127 |
| eval | 7 |
| test | 8 |
What It Covers
- Solana spot and perpetual futures research workflows.
- NVIDIA-style trading factory stages: ingestion, research, optimization, inference, execution policy, monitoring.
- RAPIDS/cuDF feature engineering and cuML KDE scenario generation.
- cuFOLIO/cuOpt Mean-CVaR optimization with leverage, budget, turnover, cardinality, and CVaR constraints.
- Vulcan/Phoenix paper strategy configs, command plans, and lifecycle guardrails.
- Rise/Phoenix read-only market data plans for exchange, market, candle, orderbook, funding, and trader state.
- Clawd perps tool-use patterns for prices, funding, order books, Jupiter quotes, paper trades, wallet checks, and risk assessment.
- Safety behavior: paper-mode default, no private keys, no front-running, no sandwiching, no market manipulation, and live execution only behind explicit gates.
Local Sources
| Path | Type | Chunks |
|---|---|---|
ai-training/trading_factory/README.md | trading_factory_workspace | 1 |
ai-training/trading_factory/solana_factory/factory.py | solana_trading_factory_adapter | 2 |
ai-training/trading_factory/solana_factory/vulcan_specs.py | vulcan_strategy_specs | 3 |
ai-training/trading_factory/solana_factory/rise_client.py | rise_readonly_client | 1 |
ai-training/trading_factory/solana_factory/cufolio_adapter.py | cufolio_optimization_handoff | 2 |
ai-training/data/strategies/strategy_manifest.json | generated_strategy_manifest | 2 |
ai-training/data/strategies/cufolio_mean_cvar_handoff.json | generated_cufolio_handoff | 1 |
ai-training/data/strategies/rise_market_data_plan.json | generated_rise_data_plan | 1 |
ai-training/data/strategies/vulcan_command_plans.json | generated_vulcan_command_plans | 1 |
ai-training/trading_factory/cufolio/README.md | cufolio_reference | 3 |
ai-training/trading_factory/cufolio/src/cvar_optimizer.py | cufolio_cvar_optimizer | 3 |
ai-training/trading_factory/cufolio/src/cvar_parameters.py | cufolio_cvar_parameters | 1 |
ai-training/trading_factory/cufolio/src/scenario_generation.py | cufolio_scenario_generation | 3 |
ai-training/trading_factory/cufolio/src/rebalance.py | cufolio_rebalancing | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/vulcan.py | autoresearch_vulcan_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/rise.py | autoresearch_rise_reference | 1 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/paper.py | autoresearch_paper_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy.md | autoresearch_strategy_reference | 0 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/ta.py | autoresearch_ta_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/grid.py | autoresearch_grid_reference | 2 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/twap.py | autoresearch_twap_reference | 1 |
ai-training/perps/functions.py | solana_perps_tools | 3 |
ai-training/perps/prompter.py | solana_perps_prompts | 2 |
ai-training/perps/schema.py | solana_perps_schema | 1 |
ai-training/perps/functioncall.py | solana_perps_agent | 3 |
ai-training/onchainai.md | onchain_ai_reference | 3 |
ai-training/README.md | training_pipeline_reference | 3 |
AGENTS.md | clawd_agent_catalog | 3 |
ai-training/data/realtime_research_dataset_manifest.json | research_dataset_manifest | 3 |
External References
| Reference | URL |
|---|---|
| NVIDIA AI Algorithmic Trading Factories | https://www.nvidia.com/en-us/use-cases/ai-algorithmic-trading-factories/ |
| NVIDIA Quantitative Portfolio Optimization Blueprint | https://build.nvidia.com/nvidia/quantitative-portfolio-optimization |
| Solizardking/cuFOLIO | https://github.com/Solizardking/cuFOLIO |
| Phoenix Vulcan CLI Strategies | https://docs.phoenix.trade/cli/strategies |
| Phoenix Rise SDK | https://docs.phoenix.trade/sdk/rise |
| Phoenix Account Health and Leverage Tiers | https://docs.phoenix.trade/phoenix/margin-and-risk/account-health |
| Solizardking/clawd-autoresearch-wiki perps | https://github.com/Solizardking/clawd-autoresearch-wiki/tree/main/perps |
Intended Use
Fine-tune a tool-use-capable instruct model, such as Hermes-3-Llama-3.1-8B, into a Solana trading-factory planner. This dataset is for research, optimization, simulation, and execution-policy training. It is not a live trading signal feed.
Safety
The dataset intentionally defaults to paper trading. It refuses front-running, sandwich attacks, wallet draining, private-key handling, sanctions evasion, and market manipulation. Live execution must be handled outside the dataset through an explicitly approved execution layer.
Source And License Notes
Generated SFT rows are released as CC-BY-4.0. Source excerpts retain their upstream attribution and licenses. The local cuFOLIO snapshot is Apache-2.0. The clawd-autoresearch-wiki perps files are treated as Solizardking project reference material for this training lane; clarify licensing before redistributing those raw source files outside the controlled training release.
Contributors
4 commits
Similar repos
README
Solana Clawd NVIDIA Trading Factory Instruct
Specialized SFT data for a Solana-native NVIDIA algorithmic trading factory. It teaches data ingestion, GPU feature engineering, alpha research, cuML KDE scenario generation, cuFOLIO/cuOpt Mean-CVaR optimization, paper execution policy, risk controls, backtesting, monitoring, and Clawd governance.
Format
Each row uses OpenAI-style messages plus metadata:
{"messages": [{"role": "system", "content": "..."}, {"role": "user", "content": "..."}, {"role": "assistant", "content": "..."}], "metadata": {...}}
Splits
Produced by scripts/prepare_dataset.py with seed 42.
| Split | Examples |
|---|---|
| train | 127 |
| eval | 7 |
| test | 8 |
What It Covers
- Solana spot and perpetual futures research workflows.
- NVIDIA-style trading factory stages: ingestion, research, optimization, inference, execution policy, monitoring.
- RAPIDS/cuDF feature engineering and cuML KDE scenario generation.
- cuFOLIO/cuOpt Mean-CVaR optimization with leverage, budget, turnover, cardinality, and CVaR constraints.
- Vulcan/Phoenix paper strategy configs, command plans, and lifecycle guardrails.
- Rise/Phoenix read-only market data plans for exchange, market, candle, orderbook, funding, and trader state.
- Clawd perps tool-use patterns for prices, funding, order books, Jupiter quotes, paper trades, wallet checks, and risk assessment.
- Safety behavior: paper-mode default, no private keys, no front-running, no sandwiching, no market manipulation, and live execution only behind explicit gates.
Local Sources
| Path | Type | Chunks |
|---|---|---|
ai-training/trading_factory/README.md | trading_factory_workspace | 1 |
ai-training/trading_factory/solana_factory/factory.py | solana_trading_factory_adapter | 2 |
ai-training/trading_factory/solana_factory/vulcan_specs.py | vulcan_strategy_specs | 3 |
ai-training/trading_factory/solana_factory/rise_client.py | rise_readonly_client | 1 |
ai-training/trading_factory/solana_factory/cufolio_adapter.py | cufolio_optimization_handoff | 2 |
ai-training/data/strategies/strategy_manifest.json | generated_strategy_manifest | 2 |
ai-training/data/strategies/cufolio_mean_cvar_handoff.json | generated_cufolio_handoff | 1 |
ai-training/data/strategies/rise_market_data_plan.json | generated_rise_data_plan | 1 |
ai-training/data/strategies/vulcan_command_plans.json | generated_vulcan_command_plans | 1 |
ai-training/trading_factory/cufolio/README.md | cufolio_reference | 3 |
ai-training/trading_factory/cufolio/src/cvar_optimizer.py | cufolio_cvar_optimizer | 3 |
ai-training/trading_factory/cufolio/src/cvar_parameters.py | cufolio_cvar_parameters | 1 |
ai-training/trading_factory/cufolio/src/scenario_generation.py | cufolio_scenario_generation | 3 |
ai-training/trading_factory/cufolio/src/rebalance.py | cufolio_rebalancing | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/vulcan.py | autoresearch_vulcan_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/rise.py | autoresearch_rise_reference | 1 |
ai-training/trading_factory/clawd-autoresearch-wiki/perps/paper.py | autoresearch_paper_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy.md | autoresearch_strategy_reference | 0 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/ta.py | autoresearch_ta_reference | 3 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/grid.py | autoresearch_grid_reference | 2 |
ai-training/trading_factory/clawd-autoresearch-wiki/strategy/twap.py | autoresearch_twap_reference | 1 |
ai-training/perps/functions.py | solana_perps_tools | 3 |
ai-training/perps/prompter.py | solana_perps_prompts | 2 |
ai-training/perps/schema.py | solana_perps_schema | 1 |
ai-training/perps/functioncall.py | solana_perps_agent | 3 |
ai-training/onchainai.md | onchain_ai_reference | 3 |
ai-training/README.md | training_pipeline_reference | 3 |
AGENTS.md | clawd_agent_catalog | 3 |
ai-training/data/realtime_research_dataset_manifest.json | research_dataset_manifest | 3 |
External References
| Reference | URL |
|---|---|
| NVIDIA AI Algorithmic Trading Factories | https://www.nvidia.com/en-us/use-cases/ai-algorithmic-trading-factories/ |
| NVIDIA Quantitative Portfolio Optimization Blueprint | https://build.nvidia.com/nvidia/quantitative-portfolio-optimization |
| Solizardking/cuFOLIO | https://github.com/Solizardking/cuFOLIO |
| Phoenix Vulcan CLI Strategies | https://docs.phoenix.trade/cli/strategies |
| Phoenix Rise SDK | https://docs.phoenix.trade/sdk/rise |
| Phoenix Account Health and Leverage Tiers | https://docs.phoenix.trade/phoenix/margin-and-risk/account-health |
| Solizardking/clawd-autoresearch-wiki perps | https://github.com/Solizardking/clawd-autoresearch-wiki/tree/main/perps |
Intended Use
Fine-tune a tool-use-capable instruct model, such as Hermes-3-Llama-3.1-8B, into a Solana trading-factory planner. This dataset is for research, optimization, simulation, and execution-policy training. It is not a live trading signal feed.
Safety
The dataset intentionally defaults to paper trading. It refuses front-running, sandwich attacks, wallet draining, private-key handling, sanctions evasion, and market manipulation. Live execution must be handled outside the dataset through an explicitly approved execution layer.
Source And License Notes
Generated SFT rows are released as CC-BY-4.0. Source excerpts retain their upstream attribution and licenses. The local cuFOLIO snapshot is Apache-2.0. The clawd-autoresearch-wiki perps files are treated as Solizardking project reference material for this training lane; clarify licensing before redistributing those raw source files outside the controlled training release.
Similar repos
Linked in READMEs
Contributors
4 commits