Soumen Halder (PhD)

AI Researcher

Mumbai, Maharashtra, India7 yrs 10 mos experience

Highly Stable

Key Highlights

Ph.D. in Experimental High Energy Physics with data analysis expertise.
Developed innovative no-code suite for financial strategy optimization.
Extensive experience in machine learning and statistical analysis.

Stackforce AI infers this person is a Fintech and Research expert with strong quantitative and analytical skills.

Contact

Skills

Core Skills

Quantitative AnalyticsSoftware Development MethodologiesQuantitative FinanceApplied Machine LearningTheoretical PhysicsStatistical Data Analysis

Other Skills

AWS deploymentAnalytical SkillsAsyncAutomated arbitrage algorithmsAutomation of processesBack-End Web DevelopmentBacktestingBusiness OperationsC (Programming Language)C++Cross-exchange tradingCross-functional collaborationData AnalysisData VisualizationData analysis

About

I hold a Ph.D. in Experimental High Energy Physics (HEP) from Tata Institute of Fundamental Research with extensive training in data analysis at KEK (Japan), where I measured some key physics observables. My academic background and hands-on experience in data analysis equip me with a unique blend of theoretical knowledge and practical skills. I am adept at applying advanced statistical techniques and machine learning algorithms to analyze large datasets, contributing to significant research advancements. In my current role, I contribute to an innovative no-code suite for strategy creation, focusing on backtesting complex trading strategies and automating processes to enhance efficiency. I also develop optimization frameworks and prototypes, collaborating closely with product and UX teams to ensure user-friendly solutions.

Experience

7 yrs 10 mos

Total Experience

7 yrs 10 mos

Average Tenure

Current Experience

Devine group

Quantitative Developer

Feb 2025 – Present · 1 yr 4 mos · [Work From Home] · Remote

Designed and deployed automated arbitrage algorithms—including cross-exchange, funding rate, and statistical arbitrage—using Binance and Hyperliquid order book data.
Modeled key microstructure dynamics: latency, bid-ask spread, slippage, and transaction costs.
Developed low-latency WebSocket collectors for real-time order book data ingestion.
Built and deployed a full-stack dashboard (frontend and backend) for 24×7 live monitoring of trading signals, executions, and system status.
Currently trading with capital; scaling planned across additional crypto pairs and exchanges.
Deployed the algorithms on AWS for continuous, fault-tolerant operation.

Automated arbitrage algorithmsCross-exchange tradingStatistical arbitrageWebSocket collectorsAWS deploymentQuantitative Analytics+1

Diracai services

Data Scientist

Aug 2024 – Present · 1 yr 10 mos · Bhubaneswar, Odisha, India · Remote

In my current role, I am contributing to an innovative project aimed at developing a world-class no-code suite for strategy creation and parameter optimization. My key responsibilities include:
Backtesting Complex Strategies: Conducting backtesting of financial strategies using trading data, incorporating technical, fundamental and sentiment data based indicators - to ensure effectiveness and reliability to ensure their effectiveness and reliability.
Automation of Backtesting Processes: Creating dynamic code to automate the backtesting process, improving efficiency and precision in strategy evaluation.
Optimization Framework Development: Designing and implementing a comprehensive optimization framework that enables users to fine-tune strategies effectively.
Prototype Development: Developing an initial prototype for both the front-end and back-end of the trading platform, facilitating user testing and feedback.
Cross-Functional Collaboration: Working closely with product development and UX design teams to ensure the platform is user-friendly and aligns with client needs.

BacktestingAutomation of processesOptimization frameworksCross-functional collaborationQuantitative FinanceApplied Machine Learning

Kek japan

Visiting Researcher

Jan 2020 – Mar 2020 · 2 mos · Japan · On-site

Contributed to 24/7 detector operation during visits in 2018, 2019, and 2020.
Monitored and reported detector performance to an international team.

Tata institute of fundamental research, mumbai

Research Fellow

Jul 2016 – May 2024 · 7 yrs 10 mos · On-site

Analyzed Peta-byte scale data from the Belle detector at KEK, Japan.
5 years of coding and data analysis experience using Python and C++.
Applied machine learning for signal-noise separation and background suppression.
Contributed to software development in the Belle II framework by writing and maintaining modules.
Conducted Monte-Carlo simulations, selection optimization, and bias studies.

Data analysisMachine learningMonte-Carlo simulationsSoftware developmentTheoretical PhysicsStatistical Data Analysis