According to a July 30, 2026 announcement carried by The Manila Times, GigaromAI launched a no-code, agentic AI trading platform designed to route retail and institutional users onto pre-built quantitative strategies with automated execution and integrated risk controls. The system is positioned as multi-asset — equities and futures are explicitly named in the release, while crypto is not specified. For the algo-trading segment, the platform enters a saturated no-code category where competition runs on accessibility rather than disclosed performance.
What the release claims
The platform packages four engineering layers. Portfolio-level risk management provides position sizing and exposure controls intended to operate across the full book rather than per trade. Security infrastructure covers upgraded encryption, multi-factor authentication, and continuous monitoring. Automated execution is described as tuned for consistency under shifting market conditions. Operational safeguards target uptime and resilience. The user-facing surface is strategy selection, not strategy construction — the core design choice is abstraction of the signal layer.
What the release does not show
No backtest statistics appear. No Sharpe ratio, no max drawdown, no slippage distribution, no latency figure, no fill-rate metric. For a product positioned as institutional-grade, the absence of disclosed execution benchmarks is the principal data point. Pre-built strategies carry two structural risks that cannot be evaluated from a press release: parameter over-fitting to the historical window used during model development, and regime drift when market microstructure shifts. Vendor-controlled signal libraries also introduce reproducibility risk — the user cannot audit the logic generating the orders.
Verification protocol before capital deployment
Five checkpoints separate deployable infrastructure from marketing. First, independent backtests across at least one full market cycle, including a drawdown regime comparable to 2022 or 2023, with results reproducible from published code or methodology. Second, a live execution audit measuring latency, rejection rate, slippage versus mid-price, and order-to-fill variance under stressed liquidity. Third, strategy-logic transparency: whether the underlying signal library is published, version-controlled, and auditable. Fourth, risk-control stress testing under correlated-asset shocks, not isolated single-position stops. Fifth, exit mechanics: withdrawal timelines, custody segregation, and counterparty exposure at the platform layer. No-code lowers the barrier to deployment. It does not lower the barrier to alpha. Until each checkpoint is satisfied, the system remains a black box with an interface.