Upwork home
Sign up
Secondary
Jonathan L.
Austin, TX

Trading Software & AI/ML Developer

Previously held the 'Top Rated' badge, but UpWork removes this badge when you haven't taken a job through their system in a while. I'm currently offering my services at a deep discount to build a good portfolio of work on here so take advanatage of that while you can. I have been engineering software for 23 years now in a number of different capacities, languages, and industry sectors. My core focus now is developing and integrating trading software whether it be for stocks, options, crypto, futures, or something else. SPECIALTIES - Trading indicators - AI or machine learning integrations - Connecting brokerage accounts to some other API or service or app - Automated financial analysis - Data pipelines - Data streams written to cloud based storage - Trading dashboards CODING LANGUAGES Python, Pine Script, MQL4 & MQL5, Java, C++, Ruby, PHP, JavaScript, HTML, CSS DATA PROCESSING Data Preprocessing, SKLearn Data Pipelines, Machine Learning Models, PyTorch, TensorFlow DATA STORAGE SETUPS AWS, Firebase, Google Colab, mySQL, MongoDB, PostgreSQL, GraphQL & Graph DBs, Excel QUANTITATIVE FINANCE / QUANTITATIVE ANALYSIS Regression Analysis, Advanced Time Series Models, Fama French Models, Principal Component Analysis(PCA) Models, Multi-factor Alpha Models, Advanced Portfolio Optimization Using Convex Equations, Portfolio Regularization KEY PROJECTS - Built a natural language processing service that scrapes financial data from SEC's EDGAR, conducts sentiment analysis, and plots changes over time. - Built a custom smart beta ETF that tracks an external benchmark. Calculated its tracking error and used that along with a covariance matrix within a convex equation to determine optimal reweighting of the portfolio to minimize tracking error. - Built a statistical significance evaluation tool for trading strategies. This tool divides the trading system's return data into quantiles for deeper analysis, and uses SciPy to conduct different hypothesis tests to evaluate if the signal’s results are from randomness or real alpha. - Built a multi-factor risk algorithm for identifying certain high-risk currency trading strategies previously deemed unusable that could be altered in small ways and made profitable in select scenarios. KEYWORDS Forex Trading, NinjaTrader, TradingView, Trading Strategy, Quantitative Analysis, Quantitative Finance, Trading Automation, Pine Script, MetaTrader 4, Pandas, NumPy, Keras, PyTorch, CvxPy, SciPy, Plotly, Beautiful Soup
Jonathan L. has more jobs. Create an account to review them

Skills

  • PyTorch
  • Deep Learning
  • Python
  • pandas
  • Machine Learning
  • Quantitative Finance
  • Quantitative Analysis
  • Jupyter Notebook

Browse similar freelancers