Param Khakhar

Associate Consultant

Mumbai, Maharashtra, India6 yrs 6 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Expert in quantitative research and trading strategies.
Proficient in machine learning and data science applications.
Hands-on experience with CNNs for predictive modeling.

Stackforce AI infers this person is a Fintech and Machine Learning specialist with a focus on predictive analytics.

Contact

Skills

Core Skills

Quantitative ResearchData ScienceMachine LearningComputer Vision

Other Skills

C++CNNCNN-AutoencoderCascading Style Sheets (CSS)Convolutional Neural NetworksData AugmentationData EvaluationData Pipeline DevelopmentData ScrapingData VisualizationExploratory Data AnalysisGitGithubHTML5Image Classification

About

Currently working on different stat-arb mid-frequency strategies across India, US, and Crypto on instruments comprising stocks, futures and options.

Experience

Alphagrep

Analyst, Quantitative Research and Trading

Jun 2022 – Present · 3 yrs 9 mos · Mumbai, Maharashtra, India

Mid-Frequency Alpha Research and Data Pipeline Development

PythonData Pipeline DevelopmentMid-Frequency Alpha ResearchQuantitative ResearchData Science

Aces-acm, iit delhi

3 roles

Vice Chair

May 2021 – Apr 2022 · 11 mos

Events Coordinator

Jun 2020 – May 2021 · 11 mos

Executive

Jun 2019 – May 2020 · 11 mos

Estee capital llc

Strategy Intern

May 2021 – Jul 2021 · 2 mos

Trained a CNN on time series data for predicting future returns for the stocks of public sector banks.
Used different trade and orderbook features for different securities and introduced new micro-price features.
Checked robustness of the predictions in terms of reproducibility and profitability over time.
Evaluated the predictions using different metrics such as net-gross profits, buy-sell accuracy, and event rates.

Convolutional Neural NetworksTime Series DataPredictive ModelingMachine LearningData Science

Jbm group

Machine Learning Intern

Jul 2020 – Aug 2020 · 1 mo · Gurgaon, Haryana, India

Trained a CNN-Autoencoder on augmented images of non-defective automobile parts.
Classified defective automobile parts from non-defective parts on the basis of reconstruction loss.
Experimented with image enhancement techniques, different losses, and regularization.
Achieved a recall of 70% and a precision of 67% for the defective parts.

Convolutional Neural NetworksImage ClassificationImage EnhancementMachine LearningComputer Vision