Data Scientist (Fresher / 1-2 Years Experience/Non IT) Full Time NEW

Prolicious is a pioneering whole food nutrition brand that aims to transform the way the world 

approaches health. Understanding that the way we eat is contributing to a significant increase 

in conditions like obesity, diabetes and PCOS, amongst others. Prolicious has developed a line 

of products that help you maintain your best possible health. We add 20% protein to 

traditional Indian dishes and snacks to offer them a fresh twist. We combine delicious flavours 

with vital nutrients by utilizing the power of whole foods like pulses, legumes, nuts, and 

seeds—ingredients you may find in your home. 

Aligned with Indian Council of Medical Research (ICMR) guidelines, our products are 

meticulously crafted in an FSSAI-certified state-of-the-art facility in Navi Mumbai. Our premier 

Curated Nutrition Program offers personalized plans crafted by experts, which help you 

achieve your health goals with ease. 

Diversity and Inclusion is one of our core values and we aim to provide equal opportunities to 

all employees and applicants. Together, we continue to build an inclusive culture that 

encourages, supports, and celebrates the diverse voices of our employees. 

We are currently seeking a data scientist to build the predictive and automation layer on top 

of existing data pipelines. The output is not just a dashboard, but a system that replaces 

manual monitoring with automated, commercial decisions.

Job Responsibilities

1. Commerce AI Agents & Automation

• Campaign & Bid Optimization: Build automated scripts to pull Amazon and 

marketplace data via API every 4 hours.

• Smart Alerting: Deploy alerts for when ROAS falls below thresholds or when “keyword 

waste” is detected (e.g., zero-purchase keywords spending >₹500/week). Inventory 

Intelligence: Monitor hero SKU inventory across Q-Commerce (Blinkit, Instamart, 

Zepto) dark stores; trigger alerts when stock falls below 7-day cover to prevent rank 

drops.

• Cost Leak Detection: Automate weekly scans of variable costs, flagging any 

discrepancies in agency invoices, fulfillment costs, or return rates.

2. Advanced Consumer Analytics

• Market Basket Analysis: Run association rule mining (Apriori/FP-Growth) on D2C data 

to identify SKU combinations for product bundling and cross-selling.

• LTV & Cohort Modeling: Build acquisition-month cohorts to track 30/60/90-day 

repeat rates and identify which acquisition sources produce the highest long-term 

value.

• Predictive Churn: Develop binary classification models to identify high-risk consumers 

for proactive CRM intervention.

3. Demand Planning & Supply Chain Intelligence

• Forecasting Models: Build time-series demand forecasts (SARIMA, Prophet, or 

XGBoost) at SKU, channel, and dark-store levels.

• Capacity Alerts: Create a system to flag when forward demand exceeds 90% of 

production capacity or when raw materials fall below 14-day cover.

• Inventory Optimization: Calculate reorder points for every node (Amazon FBA, Blinkit, 

etc.) based on lead times and demand variability.

4.Revenue Attribution & P&L Intelligence

• Unified P&L Model: Partner with Finance to build a channel-wise P&L (Net Revenue 

to Contribution Margin) that serves as the single source of truth for investment.

• Multi-Touch Attribution: Replace last-click models with Shapley-value attribution to 

correctly credit awareness versus retargeting channels.

• Pricing Elasticity: Estimate price elasticity by SKU to determine optimal discount 

depths that maintain margin.

5.Category Intelligence & NPD Support

• Competitor Monitoring: Build a repeatable methodology to size categories and track 

top competitor ASINs, pricing, and sales velocity.

• Sentiment Analysis: Deploy automated review scrapers for Prolicious and competitor 

products to extract recurring themes (taste, price, portion) and identify gaps for New

Product Development (NPD).

Technical Skills & Requirements

• Python: Proficiency in pandas, numpy, and scikit-learn for clean, version-controlled 

code.

• SQL: Experience with complex queries, window functions, and CTEs.

• Statistics & ML: Strong foundation in hypothesis testing, regression, clustering, and 

evaluation metrics (AUC, F1).

• Time Series: Expertise in seasonal decomposition and models like ARIMA or Prophet.

• APIs: Ability to handle REST API calls and JSON parsing for platform integrations.

Qualifications

• B.Sc./M.Sc. in Data Science, Statistics, or Engineering, or an MBA with a strong 

quantitative specialization.

• Open to final-year students and recent graduates