Rahul Dhangar@rahuldhangar
1
Events attended
1
Submissions made
India
10+ years of experience
About me
I am a full-stack developer, educator, cybersecurity practitioner, and lifelong learner with 10+ years of experience building software and working across web technologies, Python, Java, JavaScript, React, PHP, DevOps, and related systems. I've worked in corporate and freelance environments as a Full-Stack Developer, Senior Software Developer, and Lead Developer, and I've also spent significant time teaching CSE, BCA, and MCA students. I enjoy turning complex technical concepts into practical, hands-on solutions. My interests have expanded beyond traditional software development into AI, automation, and financial markets. I'm also pursuing an LL.B., which has given me an interesting intersection of technology, law, cybersecurity, and digital evidence. Alongside this, I've experimented with entrepreneurship and digital products, and I'm particularly interested in trading, quantitative thinking, and decision-making under uncertainty. I'm participating in the Alpaca AI Trading Agents Hackathon because I want to combine these different experiences - software engineering, AI, financial markets, and problem-solving - to build something genuinely useful rather than just another demo. As a solo developer, I am comfortable going from an idea to architecture, implementation, testing, and deployment. I am here to learn, experiment aggressively, and hopefully build something that stands out.
🤓 Latest Submissions
OptionForge - Alpaca Options Alpha Trading Agent
Most retail options trading bots suffer from two fatal flaws: deploying unhedged, static credit spreads that blow up when volatility spikes, or relying on unconstrained LLM loops that hallucinate strikes and violate margin limits. The Autonomous Options Alpha Agent introduces an institutional-grade, multi-strategy system pairing explainable LLM intelligence with uncompromising deterministic Python hard risk gates. Core Capabilities: 1. Dual-Layer Decoupled Architecture: LLMs propose strategy hypotheses; deterministic Python code strictly authorizes capital execution. Zero strike hallucinations via strict 21-character OCC formatting. 2. Regime-Adaptive Strategy Selection: Continuous Black-Scholes Greeks modeling measures 52-Week Implied Volatility Rank (IVR), IV Percentile (IVP), and ADX trend momentum across liquid index ETFs and mega-caps. Dynamically deploys Bull Put Spreads, Bear Call Spreads, and Iron Condors in high-IV regimes (IVR > 50), directional debit spreads in strong trends, and halts trading in low-volatility chop to preserve cash. 3. Hybrid Alpaca Infrastructure: Combines the Alpaca MCP Server for explainable LLM tool reasoning, the Alpaca CLI for lightweight operations, and the official alpaca-py Python SDK for low-latency WebSocket quote streaming and sub-millisecond mathematical risk checks. 4. Multi-Model LLM Gateway: Seamlessly switches between official partner Featherless AI (serverless open-source models: Qwen 2.5 72B, GLM-5.2) and Google Gemini (gemini-2.0-flash) with synthetic deterministic fallback for 100% uptime. 5. Deterministic Hard Risk Gates: Enforces 5% max risk per trade, 40% portfolio margin ceiling, 5% daily circuit breaker, 10% max drawdown, automated 60% profit harvesting, and 2.5x stop losses on a fresh $100,000 paper trading account evaluated strictly by Total Account Equity.
4 Sep 2026
