Dibyansu Diptiman profile photo 1 on Tal

Dibyansu Diptiman

Analytics & Data Science @CoinDCX | Ex-Licious | BITS Pilani

Bengaluru

3+ years of work experience

CoinDCX
Mar 20267 mosPresent

On-site

• Architected an LLM-powered Next Best Action system integrating trading screeners, market sentiment, user segmentation, and propensity models with a RAG pipeline on AWS Bedrock; projected to improve in-app retention by 10%, supporting 60% of core revenue streams. • Deployed a GenAI internal analytics platform on Databricks Genie, providing a natural language AI layer over all database tables for self-serve RCA and ad hoc querying, reducing business ad hoc requests by 60-70%. • Engineered a cross-sell propensity model (XGBoost + Markov Chain) for SIP and Earn products using 20+ behavioral and transactional features sourced from MixPanel funnels and app event data, optimizing targeting for high-propensity adopters. • Engineered a Python experimentation platform across 20+ feature flags, generating statistically significant cohorts and test-control splits; improved experimentation efficiency by 50% and powered 10+ growth campaigns.
CoinDCX
Jul 20241 yr 9 mosMar 2026

Hybrid

• Designed an NLP-based token classification system using similarity modeling and pattern matching, streamlining 400+ manually defined tags to 40+ structured categories to improve product discoverability. • Redesigned the token listing framework for Okto App, reducing listing turnaround from 6 hours to near real-time and increasing token coverage by 70%. • Validated impact of Google and Apple Sign-In on iOS through A/B experimentation, improving signup conversion by 3%.
Licious (Delightful Gourmet Pvt. Ltd.)
Jan 20231 yr 6 mosJun 2024

Bangalore Urban, Karnataka, India

• Operationalized a CLTV model with data-driven segmentation, improving customer retention by 10% and reducing loyalty plan payback period from 12+ months to 9-12 months. • Established an automated Python reconciliation pipeline for third-party financial services, saving 20 lakhs in costs, cutting manual effort by 70%, and enabling real-time transaction visibility. • Conducted Festive Sensitivity Analysis to enhance supply chain planning accuracy by 20%, predicting off-seasons and high-revenue periods to minimize waste and align inventory.
GitHub
@dibyansu24-maker

Dibyansu’s AI Stack

claude code
claude code
claude
claude
codex
codex
cursor
cursor
meta
meta
ollama
ollama

Education

BITS Pilani, Hyderabad Campus

Bachelor of Engineering, Mechanical Engineering

BITS Pilani, Hyderabad Campus

20195 yrs2024
BITS Pilani, Hyderabad Campus

Master's degree, Biology/Biological Sciences, General

BITS Pilani, Hyderabad Campus

20195 yrs2024

Join Dibyansu and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Dibyansu Diptiman on Tal

Headline

Analytics & Data Science @CoinDCX | Ex-Licious | BITS Pilani

About

Product-focused Business Analyst with a strong foundation in Data Science, working on product analytics, experimentation, and data-driven decision-making. I collaborate closely with product, business and engineering teams to turn user behavior and data into insights that improve products at scale. Experienced with Python, SQL, experimentation frameworks, Data Science models, and modern analytics tools, with a focus on building practical, end-to-end analytical & product solutions.

Location

Bengaluru

AI stack

claude code

claude

codex

cursor

meta

ollama

Work experience

Senior Business Analyst at CoinDCX, Mar 2026 - Present · 7 mos, On-site, • Architected an LLM-powered Next Best Action system integrating trading screeners, market sentiment, user segmentation, and propensity models with a RAG pipeline on AWS Bedrock; projected to improve in-app retention by 10%, supporting 60% of core revenue streams. • Deployed a GenAI internal analytics platform on Databricks Genie, providing a natural language AI layer over all database tables for self-serve RCA and ad hoc querying, reducing business ad hoc requests by 60-70%. • Engineered a cross-sell propensity model (XGBoost + Markov Chain) for SIP and Earn products using 20+ behavioral and transactional features sourced from MixPanel funnels and app event data, optimizing targeting for high-propensity adopters. • Engineered a Python experimentation platform across 20+ feature flags, generating statistically significant cohorts and test-control splits; improved experimentation efficiency by 50% and powered 10+ growth campaigns.

Business Analyst at CoinDCX, Jul 2024 - Mar 2026 · 1 yr 9 mos, Hybrid, • Designed an NLP-based token classification system using similarity modeling and pattern matching, streamlining 400+ manually defined tags to 40+ structured categories to improve product discoverability. • Redesigned the token listing framework for Okto App, reducing listing turnaround from 6 hours to near real-time and increasing token coverage by 70%. • Validated impact of Google and Apple Sign-In on iOS through A/B experimentation, improving signup conversion by 3%.

Business Analyst at Licious (Delightful Gourmet Pvt. Ltd.), Jan 2023 - Jun 2024 · 1 yr 6 mos, Bangalore Urban, Karnataka, India, • Operationalized a CLTV model with data-driven segmentation, improving customer retention by 10% and reducing loyalty plan payback period from 12+ months to 9-12 months. • Established an automated Python reconciliation pipeline for third-party financial services, saving 20 lakhs in costs, cutting manual effort by 70%, and enabling real-time transaction visibility. • Conducted Festive Sensitivity Analysis to enhance supply chain planning accuracy by 20%, predicting off-seasons and high-revenue periods to minimize waste and align inventory.

Education

Bachelor of Engineering, Mechanical Engineering, BITS Pilani, Hyderabad Campus

Master's degree, Biology/Biological Sciences, General, BITS Pilani, Hyderabad Campus

Public links

github: https://github.com/dibyansu24-maker