Lakshya Raikwar

AI-Assisted System Developer at Independent Portfolio Projects (2023-01 – Present)

Built and continuously extended a 6,000+ line algorithmic trading system through iterative AI-assisted development and prompt engineering, evolving the codebase across multiple hardening passes as real trading behavior surfaced new edge cases.
Diagnosed production bugs directly from real paper-trading logs, including a peak-R tracking bug that silently prevented breakeven stops from firing, an exchange-side stop-loss that went stale after internal trailing-stop upgrades, and an IEEE-754 float-precision bug that systematically under-sized exchange orders.
Corrected a risk-model math error where the interaction between per-trade risk percentage and the daily loss limit meant a single max-loss exit could exhaust the entire day's trading budget — an unintended interaction between two independently reasonable settings.

Algorithmic Trading System Builder at Self-Directed / Binance Futures (2023-01 – Present)

Designed a hybrid crypto trading system, paper-traded against live Binance Futures market data, implementing dual-signal confirmation (TRVM trend engine + a multi-factor Strict Scoring engine) with ATR-based position sizing.
Built a priority-ordered exit framework (breakeven, trailing stop, quick-profit, early-loss detection, stagnancy timeout, and a connection-timeout failsafe) evaluating 200–290 symbols per scan cycle, backed by a daily loss limit and per-symbol re-entry cooldowns.
Integrated Binance REST and WebSocket APIs for real-time market data and order execution. Independently prototyped an experimental local order-book engine (liquidity-wall and void detection over a live WebSocket depth stream), kept unpublished pending further validation rather than merged into the public repository. Separately diagnosed a 76-minute connectivity outage as an OS-level DNS resolution failure rather than exchange rate limiting, redirecting the fix to the correct layer.

About