navyamehta
Founding Eng. @ WOMBO AI | Ex-Jane Street, Groq, Splunk | UWaterloo CS
Waterloo, ON., Canada
Pinned Repositories
AI-credit-modelling
As a team of 4, we intend to use machine learning and artificial intelligence (with the context of statistical modelling) for social good, to evaluate credit risk and improve the access and fairness of credit to reach underserved communities.
ai-research-projects
OpenAI Gym's gaming simulation environments including Atari are used to train RL agents with Deep Q-Networks, policy parameterization and Actor-Critic Approaches to auto-play.
CFM-HACK-2018
We intend to use algorithmic trading tools, built on analysis of SMA, EMA and the golden cross, to drive portfolio analytics of TSX60 stocks.
data-sentiment-NLP
This project attempts to use a mix of traditional NLP techniques including trigrams, HNNs, and SVDs with attempts at LSTM and BERT architectures to predict and analyze sentiment projects - macroeconomic news relevance and Twitter subtext extraction.
dividend-forecast-simulation
We attempt to build predictive ML models to help investors create dividend forecasts for S&P500 companies. By building a proprietary data pipeline, we perform data preprocessing, cleaning and feature engineering on historic financial data to derive market insights.
enhancedgamedev
Enhanced C++ Game Development with Pattern Implementations
functional-system-design
I worked on developing "Faux Racket" compilers, interpreters and assemblers in Scheme to work with programs across the spectrum from simple imperative languages to assembly-level code.
jane-street-electronic-trading-comp
This project attempts to design a proprietary data processing system to algorithmically simulate equity trading strategies on Jane Street's electronic investment platform. Combining strategies ranging from pennying and MACD to arbitrage, we work towards building a healthy financial portfolio with prudent liquidity management.
optimization-algorithms
Amalgamation of intriguing research questions surrounding combinatorial optimization across domains
recommendation-engines
The repository explores varied mathematical approaches to recommendation engines over diverse user data sets.
navyamehta's Repositories
navyamehta/dividend-forecast-simulation
We attempt to build predictive ML models to help investors create dividend forecasts for S&P500 companies. By building a proprietary data pipeline, we perform data preprocessing, cleaning and feature engineering on historic financial data to derive market insights.
navyamehta/jane-street-electronic-trading-comp
This project attempts to design a proprietary data processing system to algorithmically simulate equity trading strategies on Jane Street's electronic investment platform. Combining strategies ranging from pennying and MACD to arbitrage, we work towards building a healthy financial portfolio with prudent liquidity management.
navyamehta/data-sentiment-NLP
This project attempts to use a mix of traditional NLP techniques including trigrams, HNNs, and SVDs with attempts at LSTM and BERT architectures to predict and analyze sentiment projects - macroeconomic news relevance and Twitter subtext extraction.
navyamehta/optimization-algorithms
Amalgamation of intriguing research questions surrounding combinatorial optimization across domains
navyamehta/AI-credit-modelling
As a team of 4, we intend to use machine learning and artificial intelligence (with the context of statistical modelling) for social good, to evaluate credit risk and improve the access and fairness of credit to reach underserved communities.
navyamehta/ai-research-projects
OpenAI Gym's gaming simulation environments including Atari are used to train RL agents with Deep Q-Networks, policy parameterization and Actor-Critic Approaches to auto-play.
navyamehta/CFM-HACK-2018
We intend to use algorithmic trading tools, built on analysis of SMA, EMA and the golden cross, to drive portfolio analytics of TSX60 stocks.
navyamehta/enhancedgamedev
Enhanced C++ Game Development with Pattern Implementations
navyamehta/functional-system-design
I worked on developing "Faux Racket" compilers, interpreters and assemblers in Scheme to work with programs across the spectrum from simple imperative languages to assembly-level code.
navyamehta/recommendation-engines
The repository explores varied mathematical approaches to recommendation engines over diverse user data sets.
navyamehta/signoz-template-celery-opentelemetry
Opentelemetry with Celery
navyamehta/ShortStocks
We use machine learning to customize news articles about companies to particular investing styles, to better support investor decision making and facilitate information transfer.