Business Analyst

Business Analyst F&V Intelligence, Decision Systems & AI Automation

SAVOmart | Bengaluru, On-site | Experience: 1–4 years

Build the intelligence system that powers buying, pricing, replenishment and performance decisions across SAVOmart's Fruits & Vegetables business.

The Opportunity

Fruits & Vegetables is one of the most complex categories in grocery retail. Procurement costs and selling prices change frequently, shelf life is short, demand varies by store, and thousands of decisions influence profitability and availability every day.

Across 60+ stores in Bengaluru and Chennai, SAVOmart generates data across sourcing, purchase costs, inventory, sales, pricing, margins, dump, replenishment and store performance. The opportunity is to connect these data points into a unified decision-making framework that helps teams understand what is happening, why it is happening, what action to take and whether that action worked.

We are looking for someone who can build this system end to end: from data integration and automated reporting to analytical decision engines, API-connected workflows and AI agents.

This is not just a dashboard-building or AI-agent development role.

You will build the intelligence layer that connects F&V business data to everyday operational and commercial decisions.

What You Will Build

1. An Integrated F&V Decision-Making Framework

Build a common analytical foundation across the buy side and sell side of the business.

• Connect procurement, purchase costs, vendor performance, quality, sales, pricing, inventory, replenishment, margins and dump.
• Develop a standardized data model across stores, SKUs, vendors and dates.
• Build systems that identify exceptions, explain performance changes and recommend corrective actions.
• Connect upstream buying decisions with downstream sales, margin, availability and waste outcomes.
• Establish feedback loops that capture business decisions and measure their actual impact.

The objective is to move from fragmented reports and manual analysis to a connected, evidence-based decision system.

2. Buy-Side Intelligence: Procurement & Sourcing

Build analytical systems that help the sourcing and category teams make better buying decisions.

• Analyze vendor-level purchase prices, quality, rejection rates, fill rates and supply reliability.
• Compare purchase costs with selling prices, expected margins, demand and potential inventory losses.
• Identify vendor concentration, cost variations, quality issues and procurement inefficiencies.
• Evaluate buying quantities against demand forecasts, store requirements, existing inventory and shelf life.
• Develop decision support for vendor selection, purchase quantities and sourcing allocation.
• Generate performance reports that connect procurement decisions with downstream commercial outcomes.

3. Sell-Side Intelligence: Pricing, Sales & Store Performance

Build decision systems that help the business improve sales, profitability and product availability.

• Analyze SKU-level sales, price points, margins, stock-outs and dump across stores.
• Identify where prices are too high, margins are at risk or price reductions are not delivering the expected volume.
• Measure the impact of price changes and markdowns using historical analysis and controlled experiments where feasible.
• Identify high-performing and declining SKUs, stores and store clusters.
• Develop store-specific pricing and assortment insights based on local demand and operating conditions.
• Connect sales performance with inventory availability and procurement decisions.

4. Automated Performance Reporting & Business Intelligence

Build a reliable reporting infrastructure that reduces manual effort and makes business performance actionable.

• Automate daily, weekly and monthly reports for category, sourcing, finance, DC and store operations.
• Develop drill-down reporting across date, city, store, SKU, vendor and category.
• Track key metrics including sales, gross margin, purchase cost, dump, inventory, stock-outs, pricing performance and vendor quality.
• Build exception-based reporting that highlights what changed, where it changed and what requires attention.
• Integrate data from existing dashboards, workbooks, databases, APIs and other approved business systems.
• Ensure calculations, source data and report outputs are validated and traceable.

The goal is not to produce more reports, but to make existing data easier to interpret and act on.

5. API Integrations, Workflows & AI Agents

Build the automation layer that connects data, analytical logic and business workflows.

• Develop Python-based data pipelines and integrate APIs to retrieve, transform and deliver business data.
• Use tools such as n8n to orchestrate scheduled workflows, automate report generation and route alerts.
• Build AI agents using LLM APIs such as Claude or OpenAI to interpret business exceptions, investigate relevant data and generate structured recommendations.
• Enable workflows that can retrieve information from approved systems, execute predefined analytical tools and return evidence-backed outputs.
• Deliver insights through approved channels such as email, WhatsApp or internal business applications.
• Implement logging, retries, validation, monitoring and error handling to ensure reliable execution.
• Design human approval mechanisms for consequential actions, with appropriate access controls and audit trails.

AI agents should work on top of validated data and business logic. Financial calculations, inventory constraints and critical business rules must remain deterministic and testable.

6. Decision Feedback & Continuous Improvement

Build the measurement framework that helps SAVOmart understand which decisions work.

• Record recommendations, approvals, rejections and actual actions taken.
• Compare expected outcomes with actual sales, margins, dump and availability.
• Evaluate the performance of forecasts, alerts and recommendations.
• Identify recurring errors, ineffective interventions and opportunities to improve business rules.
• Use historical outcomes and user feedback to improve the quality of future recommendations.

Every major workflow should create a measurable feedback loop between the recommendation and the business outcome.

7. Build for Scale

• Design reusable data pipelines, analytical modules and workflows rather than one-off scripts.
• Standardize configurations so the same framework can support additional stores, cities and business use cases.
• Maintain documentation, version control, testing and operational monitoring.
• Ensure the system can expand without proportionally increasing manual reporting and analytical effort.

What You Will Do

• Work directly with Category, Sourcing, Finance, Supply Chain, DC and Store Operations to identify high-impact business problems.
• Translate business requirements into data models, analytical logic, automated workflows and decision-support tools.
• Build solutions hands-on using Python, SQL, Excel and APIs.
• Integrate LLMs and workflow automation tools where they improve the speed or quality of decision-making.
• Validate data, investigate anomalies and challenge conclusions that are not supported by evidence.
• Deploy solutions, monitor their reliability and incorporate feedback from business users.
• Measure the commercial impact of the systems you build, including margin improvement, lower dump, better availability and reduced manual effort.

What We Are Looking For

Essential Skills

• Python: Data processing, analytical scripting, API integration and reusable automation.
• SQL: Data extraction, joins, aggregations and analysis across multiple datasets.
• Data analysis: Ability to structure messy data, validate calculations and identify meaningful patterns.
• Business understanding: Familiarity with sales, pricing, costs, inventory, margins, procurement or supply-chain operations.
• Automation: Experience building end-to-end workflows using n8n or similar orchestration tools.
• API integration: Ability to work with REST APIs, authentication, JSON, pagination and error handling.
• AI application development: Practical understanding of LLM APIs, structured outputs, tool calling and agent workflows.
• Problem-solving: Ability to convert ambiguous business problems into clear analytical logic and working solutions.

Good to Have

• Experience in grocery retail, quick commerce, FMCG, perishables, category management or procurement analytics.
• Experience building automated reporting systems or data pipelines.
• Familiarity with databases, Git, cloud deployment and workflow monitoring.
• Knowledge of forecasting, optimization, statistics and experiment design.
• Experience connecting multiple business systems into a common analytical framework.

We are looking for a hands-on builder who combines analytical thinking, technical execution and commercial judgment. Demonstrated ability to build working systems matters more than knowledge of any particular AI tool.