Independent builder · Brampton / Toronto

Mahimn Patel.

I build software for markets, research, and public data.

My recent work includes broker tooling, quantitative-research infrastructure, and contributions to open-source trading software. I start at the University of Toronto in September 2026, intending to study mathematics and economics.

Selected work ↓ GitHub ↗

Merged upstream NautilusTrader · reviewed Rust / Python contributions Grand Prize NSS Space Settlement Contest · led an 11-person team

2024—now Personal research and broker infrastructure

Selected work

Claims link to evidence

NautilusTrader contributions

Merged work in a Rust/Python trading engine covering IBKR startup instrument loading, benchmark-relative statistics, capture ratios, return aggregation, instrument resolution, and statistic dispatch.

Open source
2026 02

Project Saoirse

Grand Prize in the 2026 NSS contest. I led the 11-person team and wrote the political-economy chapter of a 50-page orbital-settlement proposal.

Team lead · author
2026 03

trading-algo

A personal multi-strategy research and execution system for IBKR and crypto. The interesting work is the boring work: validation, safety gates, auditability, and documenting where the backtests can still lie.

Python · ML · IBKR
2024—now 04

kite-algo

Indian-market infrastructure on Zerodha Kite: an operator CLI, idempotent order flow, reconciliation, audit logs, and tests around broker-specific failure modes.

Python · Zerodha
2026 05

worldmonitor-mcp

An MCP server and CLI for public-data services. The upstream project does the heavy aggregation; my contribution is the interface, reliability layer, and direct-API extensions.

TypeScript · MCP
Open source 06

Quirio

An AI assistant for Texas real-estate agents. A small group tried it during early access; I archived it after failing to find a repeatable way to reach users.

Early access · archived
Mar 2024—Apr 2026

How I work

Three defaults

01 · FALSIFY

Try to break the result.

Walk-forward splits, realistic costs, null tests, PBO, and Deflated Sharpe do not make a backtest true. They make it harder to fool myself.

02 · CONSTRAIN

Make dangerous actions boring.

Live orders pass explicit environment, command-line, policy, and human-confirmation gates. Automated actions remain subordinate to deterministic checks.

03 · DISCLOSE

Write down what failed.

The LLM trader was removed, the crypto book trails buy-and-hold, and Quirio was archived. Negative results are part of the work, not an embarrassment to hide.

Open source

Merged upstream

Work that someone else reviewed, merged, and now has to maintain.

Nautilus #4519Load every configured IB instrument at startup and fail closed on unresolved inputsMerged Jul 2026 ↗ Darts #3122Consistent zero-division handling across percentage and range-based forecasting metricsMerged Jul 2026 ↗ PR #4251Benchmark-relative portfolio statistics in the Rust core with PyO3 bindingsMerged Jun 2026 ↗ PR #4354Up- and down-capture ratios on the benchmark-aware calculation pathMerged Jul 2026 ↗ PR #4099Fixed cumulative-cache leakage in IBKR instrument requestsMerged May 2026 ↗ PR #4141Fixed intraday return aggregation to compound rather than sumMerged May 2026 ↗ PR #4174Added missing statistic trait methods to prevent analyzer dispatch failuresMerged Jun 2026 ↗

Selected writing

1 published

Essay 01 · 8 min · Apr 2026

Why I shelved the LLM trader.

I built a Gemini-driven trader, then removed it after the evidence failed to justify the story. Notes on the literature, the implementation, and what would change my mind. →