Autonomous research system
Autonomous Trading Agent
A research-first trading system, currently in hibernation while strategies evolve.
- Builder
- Karnveer Singh
- Status
- Hibernating
- Repository
- Private
- Scale
- Research project
Purpose
The Autonomous Trading Agent is a research-first autonomous system by Karnveer Singh that studies market behaviour, tests strategies and models risk before any autonomous execution.
It is deliberately hibernated: strategy research continues while live execution stays switched off.
Architecture
A market-data layer ingests and normalises realtime and historical data.
A strategy engine backtests and scores approaches before promotion.
A risk engine enforces exposure and drawdown limits independently of strategy logic.
Challenges
Backtest results rarely survive live conditions, so the system treats every result as provisional.
Risk controls have to be independent of the strategy that wants to trade.
Technologies
- Python
- Market data pipelines
- Risk modelling
- Autonomous agents
Research
Connected to research on autonomous agents and market intelligence.
Related research notes
- Autonomous agents: execution is the hard part
Why the difficulty in autonomous AI agents is not reasoning quality but execution, recovery and bounded authority.
- Market intelligence with AI: mapping demand, not opinions
Using AI and automation to map real demand signals — pricing, competitors, distribution and unmet needs — instead of surveys.
Frequently asked questions
- Is the Autonomous Trading Agent live?
- No. It is hibernated. Strategy and risk research continues, but autonomous execution is switched off.