---
title: "Parthasarathi Das - Data & Applied Scientist II in Greater Kolkata Area | Tal"
description: "Meet Parthasarathi Das, Senior Data Scientist @ Deloitte | Applied AI | LLMs, RAG, Agentic AI, Forecasting, MLOps, Experimentation | Python, SQL, PySpark | ISI Bangalore based in Greater Kolkata Area. AI stack includes cursor and claudecode."
image: "https://tal.club/meet/parthasarathi-das/share-image"
---

# Parthasarathi Das on Tal

## Headline

Senior Data Scientist @ Deloitte \| Applied AI \| LLMs, RAG, Agentic AI, Forecasting, MLOps, Experimentation \| Python, SQL, PySpark \| ISI Bangalore

## About

Results\-driven Data Scientist with over 6\+ years of experience delivering end\-to\-end machine learning solutions across global consulting and product environments\. Proven track record in developing scalable forecasting models, MLOps pipelines, and business intelligence dashboards that drive measurable impact\. Skilled in Python, PySpark, SQL, MLflow, Power BI, Agentic AI, LangChain, FAISS, RAG pipelines, Amazon Bedrock, multimodal LLMs, and autonomous AI systems with a strong foundation in statistics and quality management from the Indian Statistical Institute \(ISI\), Bangalore\. Experienced in working with cross\-functional teams across industries like retail, finance, media & entertainment, and supply chain, helping global clients make data\-backed decisions\. Passionate about building data products that bridge the gap between complex models and real\-world business outcomes\. Open to challenging opportunities worldwide that value analytical rigor, collaborative problem solving, and innovation in data science and AI\.

## Location

Greater Kolkata Area

## AI stack

cursor

claudecode

## Work experience

Data & Applied Scientist II at Deloitte, May 2026 \- May 2026 · 1 mo, Kolkata, West Bengal, India · Hybrid, Built an enterprise\-scale multi\-agent Comic Intelligence pipeline for a global Media & Entertainment client to accelerate live\-action screenplay adaptation workflows\. The system enabled scriptwriters to interactively query character dialogues, panel context, and narrative flows directly from uploaded comic volumes instead of manually traversing hundreds of pages\. Architected a hybrid multimodal RAG framework combining Amazon Bedrock Vision LLMs, SAM\+DINO segmentation, and MagiV3\-based character embeddings to parse comic layouts, associate dialogue bubbles with character identities, and generate hallucination\-resistant retrieval pipelines\. Implemented judge\-LLM–driven tool routing and retrieval validation for high\-precision response generation across complex narrative structures\. The solution reduced screenplay analysis effort by 10–15 hours per production workflow and lowered manual review overhead by \~3%, enabling faster adaptation cycles and scalable knowledge retrieval for production teams\.

Data Scientist at Deloitte, Jul 2024 \- May 2026 · 1 yr 11 mos, Kolkata, West Bengal, India · Hybrid, Finance Scaling & Spend Distribution Modeling – Global Media & Entertainment Client Led the development of a predictive financial planning solution to model Day 0 forecasts and forward budget projections for movie production cycles\. Implemented a Splined XGBoost model to generate accurate early\-stage \(Day 0\) spend estimates based on historical budgeting patterns and production characteristics\. Designed a hybrid forecasting framework combining XGBoost and domain\-specific heuristic models, utilizing a weighted ensemble to optimize forward\-looking budget forecasts for each film\. Achieved \~80% forecast accuracy, enabling data\-driven planning and minimizing budget volatility\. Drove a 15%\+ reduction in overspending across production budgets by improving forecasting precision and early financial decision\-making\.

Senior Consultant at EY, Sep 2023 \- Jun 2024 · 10 mos, Kolkata, West Bengal, India · Hybrid, Forecast Engine Utility \(FEU\) Project Led the design and implementation of an automated forecasting solution using Databricks, PySpark, Random Forest, and XGBoost, improving demand prediction accuracy for retail clients by over 20%\. Built a modular pipeline integrated with MLflow for model tracking and reproducibility, reducing deployment time by 40% and enabling reusability across similar engagements\. Developed a dashboard schema to reflect real\-time promotional impact on weekly sales, improving visibility for non\-technical business stakeholders\. Designed price elasticity models to help clients assess and optimize promotional pricing strategies, contributing to a 5–8% lift in revenue during targeted campaigns\.

Data Science Consultant at EY, Apr 2022 \- Sep 2023 · 1 yr 6 mos, Kolkata, West Bengal, India · Hybrid

Data Science Analyst at Louis Dreyfus Company, Jan 2022 \- Mar 2022 · 3 mos, Bengaluru, Karnataka, India, ● IMD And JAXA Rainfall Reporting and Prediction Dashboard: Built a pipeline and data\-model using T\-SQL\(SSMS\), GitLab \(Azure DevOps CI/CD\), Jenkins \(Scheduler CI\) for daily data storage and consumption\. Created a Dashboard in Power BI to populate weekly, daily, yearly rainfall reports to highlight specific areas of interest\. Currently developing a univariate time\-series model to predict upcoming rainfall

Data Scientist at Capgemini, Oct 2020 \- Dec 2021 · 1 yr 3 mos, Bengaluru, Karnataka, India, HR Analytics for Financial Services Built a predictive attrition model using Logistic Regression and Decision Trees, helping the HR team identify high\-risk employees with over 85% precision, enabling proactive retention efforts\. Integrated third\-party job market data \(via web scraping from Naukri and Monster\) with internal HR systems, improving model robustness and delivering a 20% increase in predictive accuracy\. Conducted anomaly detection and data quality audits, which uncovered key employee turnover trends and reduced reporting errors by 30%\. Financial Forecasting Model for Finance Team Transformed unstructured historical financial data into time\-series formats using ARIMA, ARIMAX, and Holt\-Winters models, improving forecast granularity and scenario planning\. Delivered forecast dashboards that supported strategic planning for quarterly budgeting, reducing manual forecasting efforts by 50% and increasing alignment with business targets\.

R&D Intern at MiQ, Feb 2020 \- Oct 2020 · 9 mos, Greater Bengaluru Area, Building Pipelines for Marketers and Advertisers to gain insight and strategize before running an ad campaign\.

## Education

Master of Science \- MS, Quality Management Science, Indian Statistical Institute \(SQC & OR\)

Bachelor of Technology \(B\.Tech\.\), Electrical, Electronics and Communications Engineering, West Bengal University of Technology, Kolkata

[View the original profile](https://tal.club/meet/parthasarathi-das)

```json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Person",
      "@id": "https://tal.club/meet/parthasarathi-das#person",
      "name": "Parthasarathi Das",
      "url": "https://tal.club/meet/parthasarathi-das",
      "image": "https://tal.club/meet/parthasarathi-das/profile-photo.jpg",
      "description": "Meet Parthasarathi Das, Senior Data Scientist @ Deloitte | Applied AI | LLMs, RAG, Agentic AI, Forecasting, MLOps, Experimentation | Python, SQL, PySpark | ISI Bangalore based in Greater Kolkata Area. AI stack includes cursor and claudecode.",
      "jobTitle": "Data & Applied Scientist II",
      "worksFor": {
        "@type": "Organization",
        "name": "Deloitte"
      },
      "hasOccupation": {
        "@type": "Occupation",
        "name": "Data & Applied Scientist II"
      },
      "homeLocation": {
        "@type": "Place",
        "name": "Greater Kolkata Area"
      },
      "alumniOf": [
        {
          "@type": "EducationalOrganization",
          "name": "Indian Statistical Institute (SQC & OR)"
        },
        {
          "@type": "EducationalOrganization",
          "name": "West Bengal University of Technology, Kolkata"
        }
      ],
      "knowsAbout": [
        "cursor",
        "claudecode"
      ],
      "mainEntityOfPage": {
        "@id": "https://tal.club/meet/parthasarathi-das#webpage"
      }
    },
    {
      "@type": "ProfilePage",
      "@id": "https://tal.club/meet/parthasarathi-das#webpage",
      "url": "https://tal.club/meet/parthasarathi-das",
      "name": "Parthasarathi Das on Tal",
      "description": "Meet Parthasarathi Das, Senior Data Scientist @ Deloitte | Applied AI | LLMs, RAG, Agentic AI, Forecasting, MLOps, Experimentation | Python, SQL, PySpark | ISI Bangalore based in Greater Kolkata Area. AI stack includes cursor and claudecode.",
      "dateModified": "2026-08-02T00:10:38.933Z",
      "isPartOf": {
        "@id": "https://tal.club/#website"
      },
      "mainEntity": {
        "@id": "https://tal.club/meet/parthasarathi-das#person"
      },
      "primaryImageOfPage": {
        "@type": "ImageObject",
        "url": "https://tal.club/meet/parthasarathi-das/profile-photo.jpg",
        "contentUrl": "https://tal.club/meet/parthasarathi-das/profile-photo.jpg",
        "caption": "Parthasarathi Das profile photo on Tal"
      },
      "inLanguage": "en-IN"
    }
  ]
}
```