Retrieve, sort and regress public data with neural nets (single NN) and chained NNS (“deep learning”) to determine optimal non-linear trading algorithms. Data to be retrieved via HTTP Curl, JSON and XHR and similar scripts. Automate data query results and input into SQL. Use Classifier and other NNs as appropriate; train, in sample, out of sample and execute actual trade positions once models are validated.
Founder is UCSB grad and 20+ years of Wall St. buy side experience. (Long/Short and Market Neutral hedged strategies).
Required: UCSB student
Operational understanding and ability to code in Python, R .Net, C++ or similar code suitable for automated datagathering from the web (structured as well as non structured).
Specific internship duties:
Evaluate, test and recommend use of simulation, risk optimization and signal software. Perform Comprehensive Backtest Analysis.
We provide decades of Wall St knowledge regarding return/risk tradeoffs, betting algorithms, etc. Commercial grade neural nets; multi processor server (8x) with several GPUs (CUDA operating knowledge helpful), 48 G of RAM etc (Win 10 OS).
Helpful but not required: advanced mathematics, statistics and probabilities.
Not necessary: Prior Wall St experience: just a strong work ethic, curiosity and a self-starter personality. Key to your success here and at any entrepreneurial firm is the ability to pick up the phone, speak to someone unknown to you and get results that allow you to efficiently proceed with the project at hand.
Learn: industry jargon, how to manage hedged positions in various financial instruments: e.g. equities & futures.
References for financial institutions, grad school etc. Also possible position as part or full time employment should research effort result in a viable product offering that generates economic institutional investment interest (hedged product offering).