Autonomous Trading Bot System
Built an autonomous multi-model trading system with a dry-run evaluation mechanism that dynamically switched to best-performing models — improving accuracy by 18% with zero downtime.
The Challenge
Financial markets are non-stationary — a model that works today may fail tomorrow. The client needed a system that could run multiple ML models in parallel and automatically switch to the best-performing one.
Approach & Solution
Developed a modular bot architecture where each bot encapsulates a different ML model (ANN, ARIMA, Prophet, ensemble methods). A supervisor module runs all models in dry-run mode, evaluates predictions against actual outcomes, and dynamically routes live capital to the best performer.
Results & Impact
- Benchmarked 4+ ML model architectures in parallel
- Dynamic model switching improved overall prediction accuracy by 18%
- Zero-downtime model rotation with dry-run validation
- Containerized AWS deployment with model versioning
More Projects
MacVis — Local-First AI Desktop Assistant
A fully-agentic macOS desktop app that gives AI full access to your Mac — bash, filesystem, web search, and MCP integrations. Bring-your-own-key: Anthropic, OpenAI, OpenRouter, Gemini, Groq, and Ollama. No backend, no telemetry, everything stays on your machine.
Full StackClaude-Powered iOS AI Assistant
Full-stack iOS AI assistant with natural-language control over device functions, Gmail/Calendar automation, AI-participated group chat, and hybrid pgvector + Supabase persistent memory system.
Full StackAI Trading Strategy Generator
End-to-end agentic backend generating 200+ trading strategy variants autonomously. Client deployed 3 profitable strategies to production with measurable revenue impact.