const frank = {
role: "Full-Stack Engineer & ML Practitioner",
location: "Earth",
focus: ["Web", "Mobile", "AI/ML", "LLM Tooling"],
building: "AI Agents · contributing to llm-exe OSS",
learning: ["Machine Learning", "LLMs", "Cloud Architecture"],
philosophy: "Ship fast, iterate faster.",
};- Going deep on Machine learning, Agentic AI development, LLM tooling, and distributed systems
- Active OSS contributor - 30+ merged PRs into llm-exe
- Ask me about Web Dev, Mobile Dev, DevOps, AWS, LLMs, or ML
- Currently building: AI-powered developer tools
|
RL Trade Execution RL agent that minimises crypto slippage on large orders. Trained on 7.5M bars of real Binance data across 3 assets, validated against 98M tick-level trades — beats TWAP by +44.1%. |
Ensemble Music Mood Classifier Finetuned ensemble combining audio feature extraction (VGG), lyrics analysis, and custom regression classifiers built in both scikit-learn and raw NumPy/Pandas — no shortcuts. |
|
Open Source Contributor · llm-exe.com 9+ merged PRs into the upstream TypeScript LLM framework — workflow automation, CI/CD hardening, release pipeline improvements, and docs-sync infrastructure. |
GitHub-Integrated Bot AI bot that summarises PRs, highlights important changes, and suggests improvements — wired directly into your repo workflow so reviews happen automatically. |



