Gaston Solari L.

Quant developer with 5+ years of experience building and operating trading systems in production environments. Strong background in data processing, event-driven architectures, and complete data/execution pipelines (Python & C++).

Led development, deployment and operation of mid-frequency trading strategies. Owned and supported infrastructure managing multi-million USD capital, including backtesting engines, data/execution frameworks, and monitoring systems. Experienced in bridging research and production, ensuring consistency across simulation, strategy validation, and live trading environments.

Jan/2024 – Jul/2025: Lead quant developer @ OTS Trading DMCC (Dubai, UAE)

• First engineer in fund’s history. Team leader of development area, and collaborator to research department. Also 2nd-in-command in account management and strategy control, for the company’s portfolio (~4.5M USD AUM, 36% CAGR).
• Designed and owned the vast majority of the trading engine. Including:

◊ [1] Data pipeline (multi-venue / exchange connectors, feed handlers, databases, etc), [2] Execution modules (order routing, OMS, RMS), [3] Strategy framework (event-driven), [4] Monitoring subsystems, and [5] User interfaces.

◊ Wire-to-wire latency (P95) in the order of 0.2-milliseconds.

◊ Execution infrastructure audited and approved by BDO’s APT global standard.

• Deployed >30 MFT strategies into production. Creator & coordinator of agile know-how process to achieve consistency across 3 stages of strategy conciliation: [1] vectorized, [2] event-driven backtesting and [3] real-time.

◊ Including: detrended mean reversions (e.g.: CMMA pairs, oscillator coupling), DSP (e.g.: ZLEMAs, Kalman, Ehlers filters), price-action pattern recognition (e.g: breakout, accumulation), order flow imbalances (from prop firm feeds).

◊ Some models used alternative data sources (web-scraping, web sourcecode inspection, message parsing)

• Later redesigned the trading engine into a low-latency trading ecosystem (based on distributed architecture), for HFT. Coded with PyBind 11 (Back-end in C++, strategies in Python)

◊ Contributed to its data-stream module including OnixS-based FIX connector (LMAX, OneZero) and WebSockets (forex & crypto exchanges).

◊ Built upon it; an event-driven backtesting module from scratch, supporting L2 order book emulation and order matching. Merge-sort tree algorithm able to execute backtests of 200M symbol-parallel ticks in < 3 minutes.

◊ Strategy framework promoted identical code reuse across backtest, validation & production, drastically improving deployment speed and out-of-sample reliability.

Apr/2022 – Sep/2023: Quant developer @ Profluent Trading INC (remote)

• Designed and deployed several HFT-monitoring systems into production, used daily by operation team and investors, including • live delay & slippage profiling of full order lifecycle (exchange tick emission → … → order ack message), • PNL and fee/rebate analysis, • and reporting/alerting in various formats (Telegram/Slack).
• Created async connectors based on WebSockets, replacing poll-based functionality (REST) by real-time streams (balance, positions, margin, PNL, K-Lines, L2 order book) across 7 crypto exchanges (see “integrations” below).
• Built unified authentication and credential storage with secure encryption (Vault), improving isolation and security of multi-venue account operations.
• Conducted large-scale data analysis on trading and execution datasets, including latency/slippage correlation, execution inefficiencies, order rejection impact, and anomaly detection in price feeds.

Mar/2019 – Mar/2022: Quant developer @ Vitrex SA (Buenos Aires, AR)

• Created a Python framework for coding event-driven (tick) trading strategies, including a ZMQ-based set of multi-account connectors linking Python with MetaTrader. Later extended to IBKR TWS and IQ Option.
• Created an external controller of MetaTrader 4's "Strategy Tester", via bash subprocesses and multi-processing pools. Backtesting of 3rd party alt-signal sources (Telegram, Discord) with real tick data from various CFD brokers.
• Conducted quant analysis and execution for options trading, consisting of put-shorting & rollover strategies. 140+ trades, 32% net return, Sharpe 1.6 over a 5-month period.
• Implemented price structure analysis tools (support/resistance, trend detection) using clustering algorithms (Gaussian Kernel Estimators, DBSCAN).

Oct/2016 – Jul/2018: Embedded Engineer intern @ ITBA (Buenos Aires, AR)

• Prototyping, consultancy and research for different robotics' projects. mostly for universities and startup initiatives. Mostly related to: datalogging, simulation, autotronics, domestic security, and mobile robotics.
• Mockup of a 14-kW Antminer-S9 Bitcoin mining installation in Rocha, Uruguay. With solar panel assemblies and liquid cooling pipes by natural convection.
• Worked in the creation and testing of solar-tracker robot with aiming algorithm based on LQG control programmed in HC08 controllers. Energy efficiency increase of ~6% in photovoltaic automation applications.
• Some tech used: Raspberry, Arduino/ESP, PLC, PCB design, CNC/G-code, Serial comm (I2C, CAM, RS232), CATIA V5, Industrial electronics

Sep/2015 – Jan/2016: CFD intern @ Candoit (Buenos Aires, Argentina)

• QA & Testing of software for Fluid Dynamics’ simulation on Oil wells

January 2020 – March 2021: EPAT @ Quantitative Learning PVT LTD (remote)

Thesis project: Event-oriented backtesting procedure for multi-currency trading strategies in Jupyter Notebook. Awarded “Project of the year” prize.

March 2014 – November 2017: Degree in Mechatronics @ ITBA (Buenos Aires, AR)

Thesis project: Complete prototyping of modular locker systems controlled with SQLite user database, Tkinter GUI and fingerprint recognition firmware in Raspberry and Arduino. Done at structural, electric and hardware level.

March 2009 – November 2017: Mechanical Engineering @ ITBA (Buenos Aires, AR)