Pranav Deepak

PD .

Analyst, developer, cinephile

Scroll
Portrait of Pranav Deepak in a sunlit cave.

Pranav Deepak

Business Analyst at Flipkart, developer by instinct.

I turn gruntwork into scripts, dashboards into decisions, and most weekends into another rewrite. Slightly addicted to automation, unapologetically.

When I am not cohortizing customers or drafting engagement strategy, I am working with ML models or wondering whether the side project needs another complete do-over. From straightforward GSheet dashboards to graph neural networks, the working solution is rarely the complex one.

Experience

From bugs to business impact.

  1. Business Analyst · Flipkart

    Leading Payment Analytics currently and led Recommendation Analytics handling analytical solutioning for Hyperlocal Replenishment and Customer Engagement with multiple Reco Systems across a variety of use cases

    • Python
    • SQL
    • BigQuery
    • PySpark
    • R
  2. SDE Intern · BNP Paribas ISPL

    Built ETL pipelines for financial data sources and handled source control migration from Visual Source Safe to Bitbucket.

    • Bitbucket
    • Microsoft VSS
    • Shell Scripting
    • Informatica PowerCenter Suite
  3. Research Intern · Sony SSUP

    Led integration team to enable microcontroller data accessibility on the cloud platform. Redesigned backend to microservices architecture and rebuilt the complete dashboard frontend

    • GraphDB Ontotext
    • MQTT
    • MERN
    • InfluxDB
  4. Application Development Intern · L&T Construction

    Developed backend Web APIs and built multiple web components for site evaluation client applications.

    • ASP.NET Core
    • Angular
    • MS SQL
    • Postman
  5. Undergraduate Student · Amrita Vishwa Vidyapeetham, Coimbatore

    Got my B.Tech in Computer Science and Engineering & found my passion for technology and design.

    • Data Science
    • Full Stack Dev
    • Computer Networks

Projects

Trial by error, heavy on the trial.

  • Music Recommendation & Analytics App

    Amusify

    Music recommendation dashboard with analytics built using Spotify API.

    • Python
    • Streamlit
    • Spotify API
  • Comprehensive Faculty Dashboard

    FacDash

    Full-stack faculty management system with worklog tracking and scheduling.

    • ReactJS
    • NodeJS
    • ExpressJS
    • MySQL
  • Information Extraction & Classification Tool

    NewsNER

    NLP-powered information extraction tool fine-tuned for Hindi news articles.

    • Python
    • NLTK
    • Spacy
    • HuggingFace
  • Customer Segmentation Analysis

    ClusterSense

    Customer segmentation analysis to identify key groups and tailor marketing strategies for a retail sales dataset.

    • Python
    • R
    • scikit-learn
    • Pandas

Side Quests

Research and work that does not fit a single label.

  • Research Publication

    Content-Agnostic Community Classification

    Meta-graph representations of conversational dynamics for community classification (OSNEM 2025 - In peer-review).

  • Hackathon

    Flipkart AnalytiQ Hackathon 2025 - Winner

    Built post-order RTO (Return to Origin) prediction system for reducing logistics costs for high order volume BUs.

  • Paper Presentation

    Annual Flipkart DS Conference 2025

    Presented seasonality and weather-based product weights for recommendation systems focusing on replenishment strategies.

  • Certifications

    Machine Learning & Deep Learning Specializations

    Completed comprehensive Coursera specializations covering ML algorithms and deep learning architectures.

Hobbies

The interests that reset my attention.

  • Piano keys and a keyboard instrument.

    Pianist

    My companion when I'm alone

  • A badminton racket and shuttlecock on court.

    Badminton

    Best sport for relaxation and focus, ngl

  • A cinema screen and auditorium representing a love of film.

    Cinephile

    Storytelling through visual narratives and cinematography catches my eye

  • A swimmer moving through a pool.

    Swimming

    A refreshing escape

Get in touch

Leave a note.

Share a little context and I will get back to you when I can.