Skip to main content
IN DEVELOPMENTOperational Trading Platform

QuantGorilla — Agentic Market Intelligence and Trading Systems Research

Johnny's role: Product architect and agentic systems orchestrator—product requirements, trading-workflow design, risk-control architecture, agent coordination, operator experience, evidence standards, and readiness decisions.

Multi-agent trading operator for Hyperliquid with scanner pipelines, risk gates, mode separation (paper/shadow/soak/live), and an operator cockpit. Research platform — not a profit guarantee.

Recruiter summary

What it is
Agentic market-intelligence and trading systems research platform with operator cockpit and mode-separated execution paths.
What I directed
Product requirements, scanner funnel design, risk gates, paper/shadow/soak/live separation, operator dashboards, and soak validation workflows across HL desk lanes.
What was hard
Designing high-stakes event-driven systems that fail safely, log rejections as first-class data, and keep operators in control.
Evidence
QuantGorilla HL repo with agentic harness evaluators and mode-labeled API/UI paths — paper and replay evidence; live promotion gated.

Product problem

Operators need agentic market-intelligence tooling that rejects most setups, never mixes paper/replay with live capital, and surfaces gate decisions — not a black-box autopilot.

Product status

In development — paper, shadow, and soak paths active; live promotion requires explicit evidence gates and operator approval.

System designed

Event-driven scanner funnel with qualification gates, mode-separated execution lifecycle (paper/shadow/soak/live), operator cockpit for health and gate visibility, and multi-lane desk architecture (HL, RH, MSTR).

  • Scanner and setup evaluation funnel with rejection logging
  • Risk-control and position lifecycle management
  • Operator cockpit with market-data health and gate visibility
  • Soak validation and readiness reporting
  • Multi-lane architecture (HL, RH, MSTR desks)

AI-assisted implementation

Directed AI-assisted implementation of scanner evaluators, agent coordination harnesses, cockpit surfaces, and soak reporting workflows — with human review on risk controls and promotion criteria.

Safeguards and evaluation

  • Mode labels enforced at API and UI layers
  • Risk budgets, kill-switch patterns, and operator override paths
  • Rejection reasons logged as first-class operational data
  • Soak and readiness reports required before live promotion
  • No investment-performance or profit guarantees in product copy

Verified evidence

  • Mode-labeled API and UI paths in QuantGorilla HL codebase
  • Agentic harness evaluators and soak validation workflows in repo
  • Operator cockpit surfaces with gate and health readouts (marketing preview may use fixture data)

Known limitations

  • Not production-ready for unattended live trading
  • Marketing previews and demos may show paper, shadow, soak, or fixture data — not live PnL
  • RH and MSTR desk lanes at varying maturity vs HL focus
  • No verified live profitability or investor-grade performance claims

Technologies used

Python/FastAPINext.jsPostgreSQLHyperliquid APITypeScriptDocker

Next milestone

Complete soak evidence package for HL desk lane and document explicit live-promotion checklist with operator sign-off.

View product surface →

Related content

Interested in this kind of product work?

I'm open to roles where AI Product & Automation Systems Specialist skills apply — agentic systems, automation, and evidence-driven product delivery.