Ankan Kar

Software Engineer

Chennai, Tamil Nadu, India1 yr experience

Most Likely To SwitchAI Enabled

Key Highlights

Expert in causal inference and statistical analysis.
Proven success in algorithmic trading with high returns.
Strong background in machine learning and data modeling.

Stackforce AI infers this person is a Backend-heavy Fullstack developer with expertise in Fintech and AI/ML applications.

Contact

Skills

Core Skills

Causal InferenceStatistical Data AnalysisAlgorithmic TradingMlopsAlgebraic ComplexityLarge Language Models (llm)Topological Machine Learning

Other Skills

AI/ML PipelineAlgorithmsAnalytical SkillsApplied ProbabilityArtificial Intelligence (AI)Audio ProcessingAudio RecognitionBackendBackend DevelopmentBashC++Causal AnalysisCausal LearningClinical ResearchClinical Trials

Experience

Parabole.ai

Software Engineer

Jun 2025 – Present · 9 mos · Princeton, New Jersey, United States · Remote

Algorithm design, data analysis, change point analysis and causal inference (PC, Greedy using BIC, AIC and other likelihoods) for predictions on batch and different variations of Granger Causality on time series data.
Building API for detection of statistical and causal strength.
Deployment of algorithms and related scripts on personal GPU servers for backtesting.
Data redundancy removal for cost optimization, identification of statistical and causal divergence.
Developing document parsing and data extraction tools using Docling and LLMs.
Mentoring Interns.

Causal InferenceStatistical Data AnalysisCausal AnalysisOptimization AlgorithmsData ModelingCausal Learning+5

Chennai mathematical institute

2 roles

Teaching Assistant

Jan 2025 – Apr 2025 · 3 mos · Chennai, Tamil Nadu, India · On-site

Data Mining and Machine Learning

Teaching Assistant

Aug 2024 – Nov 2024 · 3 mos · Chennai, Tamil Nadu, India · On-site

Design and Analysis of Algorithms

Zfense labs

Algorithmic Trading Strategist

Jun 2024 – Jun 2025 · 1 yr · California, United States · Remote

Developed backend systems for simulating and backtesting trading strategies.
Implemented classical and machine learning algorithms for HFTs.
Automated end-to-end testing workflows for trading strategies.
Integrated MLOps tools (e.g., MLflow) for model deployment and tracking records.
Best result we got in classical method is about 34% annual return and max dropdown of -11.20% with 5.24 sharpe ratio.

MLOpsQuantitative FinanceAlgorithmic TradingHigh-Frequency TradingBackendFinancial Markets+4

Indian institute of technology, kanpur

Senior Student Research Associate

May 2024 – Aug 2024 · 3 mos · Kanpur, Uttar Pradesh, India · On-site

Abstract: This project focuses into Polynomial Identity Testing (PIT) for Read-Once Algebraic Branching Programs (ROABPs) and Log-Variate circuits. We systematically explore the development of PIT ideas in these models and their efficiency and unique properties. ROABPs, a specialized form of Algebraic Branching Programs (ABPs) where each variable appears only once per path, allow for potentially more efficient PIT algorithms due to their structured nature. This report also investigates a
conjecture presented in [3], applying various methods and tried to examine its validity. Additionally, we provide an overview of Log-Variate circuits, highlighting their relevance in the context of PIT. All proofs not included in this paper can be found in the cited references.

Algebraic Complexity

Coriolis technologies pvt. ltd.

Summer Intern

May 2024 – Jun 2024 · 1 mo · Pune, Maharashtra, India · Remote

• Developed a local copilot using open-source LLMs to analyze private codebases for code autocompletions.

Search AlgorithmsLarge Language Models (LLM)Backend

Chennai mathematical institute

Teaching Assistant

Jan 2024 – Apr 2024 · 3 mos · Chennai, Tamil Nadu, India · On-site

Algorithms and Design Techniques

Parabole.ai

Research Intern

May 2023 – Feb 2024 · 9 mos · New Jersey, United States · Remote

Generated synthetic datasets for evaluating causal inference models and Enhanced structural learning.
Designed changepoint detection technique to identify regime shifts in causal structures.
Constructed causal models directly from observational datasets.

Causal InferenceOptimization AlgorithmsCausal LearningBackend Development