John is currently unavailable
John Lazar
Verified Expert in Engineering
AI Developer
London, United Kingdom
Toptal member since December 30, 2019
Bio
John is a senior quant developer and AI developer who's worked the full development lifecycle of various applications. He's proficient in several languages—C++, C#, Java, Python, and R, to name a few—across Windows and Linux. The types of projects where John excels are advising and implementing in-house cloud distributions of heavy-load operations combined with continuous integration and deployment (Linux, DevOps packages).
Portfolio
Wintermute
Rust, Python, Crypto, Decentralized Finance (DeFi)
AQAI Capital
C++, Python, MetaTrader 5, Automated Trading Software, MQL5, Financial Data...
Stanford University
PyTorch, Deep Learning, Machine Learning, Time Series Data
Experience
- Algorithmic Trading - 15 years
- Python - 10 years
- C++ - 9 years
- Market Risk - 7 years
- FX - 6 years
- Credit Risk - 4 years
- Machine Learning - 4 years
- Commodities - 1 year
Work Experience
DeFi Developer
2025 - PRESENT
Wintermute
- Optimized AMM arbitrage strategies based on competitors' market behavior. Evaluated AI automated parameters adjustment.
- Created/improved several DEX exchanges integrations (Phoenix, Pumpfun, Bluefunpro, Hyperliquid, etc).
- Contributed to improving the core infrastructure latency.
- Handled gas estimation for AMM swap smart contracts across various volumes, to be utilized in AMM arbitrage strategies, on EVM chains and Solana.
Technologies: Rust, Python, Crypto, Decentralized Finance (DeFi)
Senior Developer
2024 - PRESENT
AQAI Capital
- Designed genetic algorithm trading strategies on a large number of symbols with good results in C++ for implementation on MetaTrader 5 (MT5) brokers.
- Developed a high-speed backtesting system to be used by the genetic algorithm in C++.
- Built MT5 and MT4 integrations as expert advisors connecting with the C++ code via a DLL, using Redis and ZeroMQ for communication.
Technologies: C++, Python, MetaTrader 5, Automated Trading Software, MQL5, Financial Data, Time Series Data
AI Course Facilitator
2023 - PRESENT
Stanford University
- Taught students AI, machine learning, and deep learning.
- Provided office hours, graded assignments, and assisted and supported new learners in getting comfortable with AI.
- Consulted on the following courses: XCS221: Artificial Intelligence: Principles and Techniques, XCS330: Deep Multi-Task and Meta Learning.
Technologies: PyTorch, Deep Learning, Machine Learning, Time Series Data
ML Engineer | Options Trading
2024 - 2024
Gains LP
- Developed a reinforcement learning (RL) trading strategy using options and volatility surface time series.
- Created an RL agent with multiple sub-agents in a multi-agent environment.
- Utilized RLlib and GPU to improve the trading system training time.
Technologies: Machine Learning, Trading, Jupiter, Python, Google Colaboratory (Colab), Reinforcement Learning, Options Trading, Option Pricing, Binary Option Trading, Futures & Options, Financial Data, Time Series Data
Senior Financial Quantitative Engineer
2024 - 2024
Umoja Labs
- Developed a synthetic options algorithm that replicates put and call options with low costs.
- Managed a Python-based crypto trading algorithm and on-the-spot and perpetual options.
- Worked on a complex system with multiple interconnected APIs for the UI back end and treasury algorithm.
- Added regression testing with mocking to simulate a complex set of interactions.
Technologies: Quantitative Finance, Python, Fintech, Decentralized Finance (DeFi), Cryptocurrency, Blockchain, Financial Data, Time Series Data
Senior Software Developer
2023 - 2024
SEFE Marketing & Trading
- Built Japan Stack and S&D models in Python, based on Excel existing implementations.
- Created a predictive model for Japan LNG imports, based on sources/scrapes.
- Processed data and created reports for commodities supply and demand models based on weather for European countries.
Technologies: Python, Machine Learning, PyTorch, MongoDB, SQL, Financial Data, Time Series Data
Machine Learning Expert
2023 - 2023
Visual Candy Systems Limited
- Developed a predictive AI model based on equity time series data.
- Utilized time series prediction based on deep learning, which is based on transformers and attention (the building blocks of LLMs and ChatGPT).
- Created a good trend entry signal and worked on developing it into a trading strategy utilizing reinforcement learning (RL).
Technologies: Scikit-learn, Machine Learning, Python, Trading, Automated Trading Software, MQL5, Financial Data, Time Series Data
Data Scientist
2023 - 2023
SixPoint Capital Management LLC
- Converted a complex Excel debt simulation calculator to an HTML/JavaScript implementation while keeping a high code quality, based on an OOP structure taking inspiration from Pandas in Python.
- Utilized custom Plotly charts to cover all aspects needed by the client.
- Guided the client to a cost-efficient solution, avoiding unnecessary hosting and DevOps costs and keeping high responsiveness of the calculator.
Technologies: HTML, JavaScript, Machine Learning, MySQL, Python, Data Science, Data Analytics, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Principal Software Developer
2020 - 2022
Ordaos Bio
- Participated actively in the small team that used an AI model to create an antibody neutralizing COVID-19 Omicron and other variants.
- Contributed to the design of an AI multi-task learning model based on attention for understanding proteins based on sequence and structure. The models were done from scratch, similar to BERT (MLM) and next-token prediction models.
- Worked on a protein generator model based on Reinforced Learning targeting generation according to a diverse set of criteria.
Technologies: Python, Machine Learning, Artificial Intelligence (AI), Algorithms, Predictive Modeling, Neural Networks, Data Science, Statistics, OpenAI, Generative Pre-trained Transformer 3 (GPT-3), Deep Learning, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), SQL, PyTorch, TensorFlow, Amazon Web Services (AWS), Google Cloud Platform (GCP), Python 3, Language Models, Google BigQuery, Programming, PySpark, Analytics, Data Engineering, Azure DevOps, Azure, Azure Machine Learning, AI Design, Architecture, Fine-tuning, Causal Inference, Reinforcement Learning, Databases, APIs, CI/CD Pipelines, Containerization, Pytest, Selenium, gRPC, Docker, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Quant Developer
2018 - 2019
BNP Paribas
- Worked as part of the credit and FX teams—migrating and enhancing existing risk and scenarios.
- Used C++ and ADA to adapt code to the new architecture, market data conversion, and new credit scenarios.
- Developed a plugin/service for the market data library of BNP, for rolling credit curves from close market to open market. It was written in Go and interacted with other services written in C++ and C#.
- Created a web pivot visualization for large datasets (bigger than the typical Excel size), based on jQuery, with two back-end versions for data processing (written in C#, Python, Pandas, and NumPy for the reconciliation of PVs between old and new pricing systems).
- Developed the Linux version of the quant library, ensuring the reliable build and testing in TeamCity along with a smooth operation and review of issues related to deployment.
Technologies: TeamCity, Linux, NumPy, Pandas, Python, jQuery, HTML, Risk Models, FX, Go, C#, C++, Algorithms, Statistics, SQL, Quantitative Finance, Finance, Programming, Analytics, Bash, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Containerization, Selenium, gRPC, Docker, Software Architecture, Trading, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Quant Developer
2017 - 2018
Lloyds Banking Group
- Evaluated GPU options and third-party libraries for various algorithms of parallelization utilized in the Monte Carlo valuations of trades.
- Created a set of KPIs that assesses the behavior of the quant models and the stability of the end of day batch. The project was done in Python using big data technologies (Pandas/Spark), generating reports in LaTex.
- Developed the market data processing system on the Bloomberg and Reuters feeds for equity indices, stocks and FX volume surfaces. The project was written in C#, based on a stateless parallel acyclic graph calculation.
Technologies: LaTeX, HTML, Pandas, Python, NVIDIA CUDA, C#, C++, Algorithms, Statistics, SQL, Quantitative Finance, Finance, Programming, PySpark, Analytics, Data Engineering, Bash, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Containerization, Selenium, Docker, Software Architecture, Trading, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Quant Developer
2016 - 2017
BNP Paribas
- Wrote the pricing for a number of products, some having simple mathematical descriptions, e.g., exotic cap floors while others requiring calibration and Monte Carlo simulations.
- Worked on the market data library (in Go), on adapting the market data necessary for the products.
- Ported modules of the quant library from Windows to Linux. As a result of the porting reduced compilation times from 30 minutes to three minutes.
- Added a methodology for coding under Linux (using JetBrains CLion and CMake).
Technologies: TeamCity, Linux, Risk Models, Go, C#, C++, Algorithms, Statistics, SQL, Quantitative Finance, Finance, Python 3, Programming, Analytics, Bash, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Trading, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Quant Developer
2015 - 2015
Credit Suisse
- Designed a scenario definition framework for CCAR and PRA, based on macroeconomic factors, using F# and C#.
- Worked on the translation of the high-level scenarios to specific bump parameters required as input by the risk systems.
- Evaluated CCAR and PRA internal implementation based on macro-economic factors vs a third-party library providing conversion to specific bump parameters.
Technologies: F#, C#, Algorithms, Statistics, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Quant Developer
2012 - 2015
Barclays
- Worked as part of the FX team on a real-time pricing system capable of handing volatile events, lack of input from primary markets. The system attempted a small prediction capability based on the order book. Worked in Java, KDB, and R.
- Developed FX real-time execution algorithms based on price, order book, volume and time. Written in Java, KDB, MySQL.
- Analyzed a large historical order book and tick data, from a KDB database, using Q queries, combining data from HDB with recent data from RTDB.
- Wrote a trade translation system from XML to internal Barclays format and self-documenting the exact translations as HTML and LaTex.
- Enhanced and refactored a legacy risk calculation system for commodities traders, providing PV and greek values.
- Created a distributed grid for the quicker processing of results.
Technologies: LaTeX, HTML, C#, C++, MySQL, Kdb+, R, Q, Java, Algorithms, Statistics, SQL, Quantitative Finance, Finance, Programming, Bash, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Team Leader
2011 - 2012
Mitsubishi UFJ Securities
- Managed a team of five people and was in charge of the timely delivery of new products.
- Implemented a QuIC pricing and risk system (currently called Markit Analytics).
- Worked on IR, FX, credit, and equity products by mapping data from Murex—incoming MXML and various database queries—to the QuIC internal format.
- Ensured pricing consistency by implementing Excel replication of all the formulas from both Murex and QuIC.
Technologies: XPath, XML, C#, Algorithms, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Team Leader | Senior Developer
2010 - 2011
Lloyds Banking Group
- Managed a team of four developers in charge of a trade storage solution based on SQL and Coherence for the quick retrieval of T0 and T-1 for pricing and trade blotters clients and regular retrieval times for historical as-of-date requests.
- Worked on a C#/MSSQL server platform, responsible for holding data from many systems in the bank, with the ability to define and manage new types of data storage in a fully dynamic way.
- Evaluated Coherence distributed cache capabilities for storing large sets of data.
Technologies: Microsoft SQL Server, Coherence, Java, C#, Algorithms, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Developer
2010 - 2010
Citibank
- Developed new products in the MARS risk system that was a Java-based system using a proprietary XML workflow.
- Combined trade data from Gemfire with market data and ensured trade execution on a distributed Symphony grid.
- Connected the pricing systems via JMS to various risk results systems.
Technologies: Java Message Service (JMS), XPath, XML, Java, Algorithms, Quantitative Finance, Finance, Programming, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Developer
2009 - 2009
Credit Suisse
- Developed a new risk platform for exotic trade books. Written in C#, it also uses grid computing (DataSynapse equivalent), GemFire trade storage, Tibco EMS communication, and OLAP for storing and allowing quick slicing/drill down on results.
- Enhanced the existing quant library pricing architecture for more stability on exotic, complex trades.
- Created various custom scenarios for risk, as required by the traders.
Technologies: OLAP, TIBCO Enterprise Message Service (EMS), C#, Algorithms, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Developer
2008 - 2009
BNP Paribas
- Developed an FX auto-execution system (C# with Tibco RV connectivity and a DevExpress Windows UI) for the real-time execution of large orders during volatile market conditions. Won praise from the head trader on algorithmic FX trading for its correct execution during the Non-Farm Payroll event.
- Built an order management application in C# with a Windows UI based on DevExpress.
Technologies: DevExpress, TIBCO Rendezvous, C#, Algorithms, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Senior Developer
2007 - 2008
Credit Suisse
- Developed a credit risk system (C#) that converted Excel EOD spreadsheets.
- Integrated the quant libraries, market data, and internal grid calculation.
- Led a team of two developers in Pune, India, to further develop the risk system.
- Created a CDO tranches substitution model for evaluating changes in the CDO tranches as PV and Delta.
Technologies: C++, C#, Algorithms, SQL, Quantitative Finance, Finance, Programming, .NET, Architecture, Databases, APIs, CI/CD Pipelines, Software Architecture, Quality Assurance (QA), Quality Control (QC), Financial Data, Time Series Data
Project Manager | Lead Developer
2005 - 2007
LSI Soft
- Managed a team that developed an automated FX trading platform for a group of traders in Denmark.
- Completed a full emulation of Metatrader (written in C# using the MS SQL database) which enabled tick data backtesting and genetic algorithms to determine the parameters of the best trading system.
- Created a web interface for trading signals publishing. Written in C#, using ASP.NET and Microsoft SQL.
- Developed multiple CRM systems for different companies/industries. Written in C# and VB using Oracle or Access as databases.
- Created a document management system, having a database of scanned and OCR-ed documents. The project was developed in C#, ASP.NET, HTML, JavaScript, Ajax dynamic loading, and an Oracle database.
Technologies: Microsoft Access, Microsoft SQL Server, MetaTrader, C++, C#, SQL, Programming, .NET, Architecture, Databases, Software Architecture, Quality Assurance (QA), Quality Control (QC), Automated Trading Software, MQL5, Financial Data, Time Series Data
Project Manager | Lead Developer
2002 - 2005
Prier VJ Software Solutions
- Managed a team of 12 developers working on a ERP project for the transportation industry.
- Contributed to the requirements gathering for all aspects of the transportation sector—from HR, drivers, trucks, pallet freight, maintenance, etc. The ERP was written in Java, using EJB.
- Worked on other projects in C#, using ASP.NET, HTML, and JavaScript.
Technologies: JavaScript, HTML, C#, SQL, Programming, .NET, Architecture, Databases, Software Architecture, Quality Assurance (QA), Automated Trading Software, MQL5
Developer
2000 - 2001
Scala Business Solutions
- Designed and developed (using C# and MS SQL) a tax management system for local administration, with modules for fees and taxes or houses, cars, land, commercials, and more.
- Built a timesheet management application (using ASP, VB, and MS SQL) for submission and approval, with companies, agencies, and freelancers as users.
- Developed multiple websites using ASP, HTML, and JavaScript.
Technologies: JavaScript, VB, Active Server Pages (ASP), C#, SQL, Programming, .NET, Architecture, Databases, Software Architecture, Commercials
Web Developer
2000 - 2000
IPACRI
- Created a web interface for financial risk tools for the client, Group ELSAG Italy.
- Implemented dynamic web loading for a table so that the data in table rows were generated by the server every 10-20 seconds.
- Integrated ASP using VB scripting with COM components (written in C++ and MS SQL database).
Technologies: Microsoft SQL Server, VB, Active Server Pages (ASP), HTML, C++, SQL, Programming
Web Developer
1999 - 2000
Webcam, Ltd.
- Developed multiple HTML/CGI web applications using C++ and Perl under Linux.
- Worked on the first version of shopping carts and used files for data storage.
- Created WAP interfaces for mobile devices with small screens.
Technologies: Wireless Application Protocol (WAP), CGI, Perl, C++, SQL, Programming
Project History
Arduino PWM Frequency Generator
An Arduino-based PWM frequency generator with commands in a format similar to GCode, which can be provided via serial interface or Bluetooth connectivity to a phone or computer.
We designed it to be an inexpensive, completely decoupled signal generation module that operates from a battery and is ready to be plugged into an IGBT amplifier for an electronics workbench. It operates via Bluetooth—with two-way communication—enabling isolation from the main computer and is useful while working with higher voltage and power applications.
We designed it to be an inexpensive, completely decoupled signal generation module that operates from a battery and is ready to be plugged into an IGBT amplifier for an electronics workbench. It operates via Bluetooth—with two-way communication—enabling isolation from the main computer and is useful while working with higher voltage and power applications.
Algo Trading Time-series Prediction Model
Worked on a time-series prediction model, recently with AI, based on raw tick data from multiple sources, over FX, indices, equity, and cryptocurrencies.
Reprocessed tick data to extract multiple sets of features. Implemented AI attention-based models and RL algorithms to generate good predictions.
Reprocessed tick data to extract multiple sets of features. Implemented AI attention-based models and RL algorithms to generate good predictions.
Education
1999 - 2004
Master's Degree in Computer Sciences
Politehnica University of Bucharest - Bucharest, Romania
Certifications
JUNE 2011 - PRESENT
PRINCE2
APMG
Skills
Libraries/APIs
PyTorch, TensorFlow, PySpark, Interactive Brokers API, jQuery, NumPy, Java Message Service (JMS), Pandas, Scikit-learn
Tools
*nux Shells, Pytest, Azure Machine Learning, LaTeX, TIBCO Enterprise Message Service (EMS), TIBCO Rendezvous, Microsoft Access, TeamCity, Jenkins, IncrediBuild, Murex
Languages
C#, C++, Python, SQL, Python 3, MQL5, Bash, HTML, F#, XML, XPath, JavaScript, Active Server Pages (ASP), VB, Perl, Java, R, Q, Go, Rust
Frameworks
.NET, Selenium, gRPC
Platforms
Linux, Docker, MetaTrader 5, Ubuntu, Azure, NVIDIA CUDA, MetaTrader, Amazon Web Services (AWS), Google Cloud Platform (GCP), Blockchain
Storage
Databases, MySQL, Microsoft SQL Server, Kdb+, MongoDB
Paradigms
Azure DevOps, OLAP
Industry Expertise
Trading Systems
Other
Risk Models, Market Risk, FX, Credit Risk, Commodities, Algorithmic Trading, Machine Learning, Artificial Intelligence (AI), Algorithms, Predictive Modeling, Neural Networks, Generative Pre-trained Transformer 3 (GPT-3), Deep Learning, Natural Language Processing (NLP), Quantitative Finance, Finance, Language Models, Programming, AI Design, Architecture, APIs, CI/CD Pipelines, Containerization, Generative Pre-trained Transformers (GPT), Software Architecture, Quality Assurance (QA), Quality Control (QC), Automated Trading Software, Financial Data, Time Series Data, Data Science, Statistics, Elliptic Curve Cryptography, OpenAI, Forecasting, Analytics, Reinforcement Learning, Trading, Risk, Coherence, DevExpress, CGI, Wireless Application Protocol (WAP), Crypto, Equities, Google BigQuery, Data Engineering, Fine-tuning, Causal Inference, Data Analytics, Commercials, Fintech, Decentralized Finance (DeFi), Cryptocurrency, Jupiter, Google Colaboratory (Colab), Options Trading, Option Pricing, Binary Option Trading, Futures & Options
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring