I build AI copilots, intelligent analytics, and automation that ship.
Pragmatic systems, clean interfaces, measurable outcomes.

Recent work: LLM workspace audits • consensus draft analytics • live‑market valuation pipelines

What I Do

• AI copilots that take action (multi‑provider, safe fallbacks, observability)
• Decision‑support analytics (aggregation, ranking, reporting, exports)
• Production ML/MLOps (containers, CI, metrics, reproducible pipelines)

Selected Work

• Problem → Fragmented pages, inbox bloat, duplicative databases slow teams.
• Approach → Multi‑provider LLM orchestration (Claude/GPT/Gemini), deep health metrics, JSON reports.
• Impact → Actionable playbooks; cached runs cut API spend significantly.
• Docs → See case study below.

• Problem → Fragmented rankings and noisy advice; hard cross‑positional calls.
• Approach → Aggregates 5+ sources, VBD (VOLS/VORP/BEER), presets, offline caching.
• Impact → Calm draft day with reproducible boards + Sheets/CSV exports.
• Docs → See case study below.

• Problem → Ad‑hoc valuations lack consistency and auditability.
• Approach → Live data (Yahoo, FRED), parallelized DCF, risk/growth adjustments.
• Impact → Investor‑ready CSV/JSON outputs for dashboards and research.
• Docs → See case study below.

Operating Principles

• Speed with safety: fast loops, strong guards, reversible changes.
• Observability first: logs, metrics, JSON artifacts; decisions are inspectable.
• Simplicity scales: clear boundaries, predictable deployments.
• User empathy: defaults that work, frictionless installs, readable reports.

Writing & Resources

• Academy Orientation — docs/academy/orientation.md
• Command‑Line Toolkit — docs/playbooks/command-line-toolkit.md
• IDE & Environment Verification — docs/playbooks/ide-setup.md
• Mathematics for AI — docs/foundations/math-for-ai.md
• Python Essentials — docs/foundations/python-essentials.md

Let’s Build

• Email: [email protected]
• LinkedIn: linkedin.com/in/brandonpshay/
• Open to full‑time roles and selective consulting.