.png&w=256&q=75)
Albert Alcance@albertalcance99145
5
Events attended
2
Submissions made
Philippines
2+ years of experience
About me
I am an AI enthusiast and creative technologist exploring how emerging technologies reshape the way we work, create, and live. Powered by AI, driven by imagination, and proven by results, I harness breakthrough language models and generative AI to create short films and music through synthetic media—blending story, sound, and visuals that captivate audiences while driving measurable commercial success.
🤓 Latest Submissions
GlassBox AI Quant
GlassBox AI Quant is an autonomous, news-driven options trading agent built on Alpaca’s developer stack. Its strategy is simple: compare the move implied by news with the move the options market is already pricing. Alpaca’s News API streams headlines in real time. A deterministic filter removes stale, repetitive, or weak stories before any model is used. Survivors go to an LLM that returns schema-validated estimates of expected move, horizon, confidence, and materiality. The LLM never places orders. GlassBox is driven by magnitude, not directional prediction. It compares the forecast move with the ATM straddle’s implied move and adjusts for movement that may already have occurred. If expected movement is materially above implied, it buys defined-risk convexity. If materially below, it sells defined-risk premium. If they broadly agree, it does nothing. Entry discipline is restrictive. A signal must survive sizing, liquidity, session timing, Greeks, concentration, correlation, macro-event, corporate-action, and other checks. The gate contains 19 named vetoes; any one can refuse the trade. Positions are defined-risk, submitted as atomic multi-leg limit orders, with deterministic IDs preventing duplicate fills. Models may influence selection and size, but deterministic code owns the limits: 1% base risk, 1.5% maximum loss per position, 6% portfolio heat, and bounded net delta. A separate supervisor can flatten and halt the system if the trader process fails. Open positions are managed through an ordered 11-barrier exit ladder covering deadlines, expiry and macro risk, thesis completion or break, stops, bell decisions, trailing protection, breakeven, profit, and time. Every decision, including every refusal, is written to a hash-chained audit log. Alpaca MCP and CLI provide independent inspection, while the dashboard shows why the system traded, why it refused, and which rule made the decision.
4 Sep 2026
LexPar AI
The Problem Independent lawyers handle most litigation but can't afford mock juries or senior-partner review. They walk into court never having heard their argument survive live, adversarial pushback — and have no way to prove trial-readiness to win premium clients. The Solution LexPar AI is a voice-driven courtroom flight simulator. Privately, an attorney spars against an AI Opposing Counsel that cross-examines and objects mid-sentence, then an AI Judge delivers a spoken ruling and scorecard. Publicly, it's a client-conversion tool: a solo attorney runs a live 10-minute session in front of a prospect to prove preparation and justify a premium retainer — a sales motion no rehearsal tool is built around. Architecture & AMD Advantage A React frontend streams audio over self-hosted LiveKit WebRTC to Python workers and a FastAPI backend. Opposing Counsel's reasoning model, Qwen2.5-72B, runs on a single AMD Instinct MI300X via vLLM/ROCm; its 192GB of HBM3 fits the 72B model plus case context on one GPU (~174GB resident, 100% GPU under load). A provider-agnostic router makes inference a config switch, not a code fork: one env var moves Opposing Counsel between cloud and the on-box MI300X, one-line rollback. Three systems keep the loop live and grounded: a 3-tier objection classifier fires instantly on clear infractions via text matching; replies are verified sentence-by-sentence against a structured ledger — gating fabricated citations before ElevenLabs TTS, cutting time-to-first-audio ~50%; and dual-corpus RAG grounds every argument in the pleadings and verbatim court rules, with a per-ruling citation audit trail. Commercial Potential LexPar targets the B2B legal market. Closed APIs bill per-token for hours of continuous voice; open-weight models on AMD run on flat compute — a high-margin SaaS model, with privileged case files that never leave owned infrastructure.
13 Jul 2026
