Portfolio index

Navigate Bongo Seakhoa

Direct downloads

Download Resume PDF Download Curriculum vitae PDF

Decision-system engineering

FXPM: validation and execution

A production-focused trading research system combining regime-aware strategy selection, stateful optimisation, statistical validation and broker-aware execution controls.

Professional system

Project overview

A production-focused trading research system combining regime-aware strategy selection, stateful optimisation, statistical validation and broker-aware execution controls.

Problem

A promising historical strategy is not sufficient evidence for reliable live execution. Selection, transaction costs, changing regimes and broker constraints must be evaluated together.

Engineering scope

The reviewed development baseline spans a 50-strategy pool and four market regimes, with persisted optimisation state, execution checks and a companion dashboard.

Role: System design, validation research and AI-assisted engineering

Context and constraints

  • Research outputs must remain separate from evidence of funded live performance.
  • Execution depends on broker contract metadata, available market data and explicit readiness decisions.

Technology stack

Technology stack
  • Python
  • Numba
  • Optuna
  • MetaTrader 5
  • Statistical validation

Contributions

  • Developed regime-aware candidate selection and persisted optimisation state rather than treating each run as a disconnected experiment.
  • Connected winner evidence, no-trade outcomes and fail-closed readiness checks to the execution path.
  • Investigated validation-gate design, dependence, selection effects and evidence sufficiency as part of continuing research.

Validation and outcome

  • The development implementation distinguishes tradeable winners from paper-only, blocked and no-trade states.
  • Execution checks cover broker preflight, position sizing, spread quality and margin protection.
  • Research examines simulation and historical evaluation separately, including the limits of observed results.
Outcome

An integrated development and research system that makes validation and execution constraints explicit. Current work focuses on the strength of selection evidence and reliable live behaviour.

Evidence scope

  • Source-reviewed development architecture and client-safe research summary, 8 September 2026.
  • The linked research record reports acceptance at ISM 2026, not a published proceedings citation or audited trading result.

Public scope

Private implementation and confidentiality boundaries limit the detail shared here. This overview focuses on the system's purpose, Bongo's role and the technologies used.

Constraints and limits

  • No profitability, return, win-rate or production-readiness claim is made here.
  • Historical performance and implemented safeguards do not guarantee future results. Strategy selection and calibration remain active research.

Continue exploring

Professional system AI context and governance system

MetaPOS Mind

Role: Primary developer, AI context and governance engineering

A Python and MCP-compatible context system that turns permission-scoped company knowledge into reliable working context for employees and AI assistants.

Technology stack
  • Python
  • MCP
  • JSON Schema
  • YAML
  • PyYAML
  • GitHub Projects
  • AI-assistant integration
Professional system Governed AI systems

OmniMind

Role: System design and AI-assisted implementation

A local AI-work control plane that compiles bounded, task-specific context and coordinates evidence, provider handoffs and human-approved capabilities without duplicating project data.

Technology stack
  • Python
  • MCP
  • Context engineering
  • Evidence gates
  • Capability routing