The client was one of the world's largest natural resources conglomerates.

### **VISION**

Digitalisation of key business units was the primary focus for client. Through it, they wanted to transform into a data-driven organisation enabling improved recovery, optimised production, enhanced safety and maximised productivity.

### **CHALLENGES**

*   Critical business data residing across multiple sources with low visibility.

*   \~50% of time consumed in data collection, aggregation and presenting to stakeholders. Delay in availability due to manual efforts.

*   Lack of a platform to view different scenarios and identify the best ones for enhanced decision-makking

*   Low visibility to management for critical consumables consumption and procurement. Manual enquiries led to delays in action.

### **OBJECTIVES**

*   Setting up a baseline framework for powerful data analytics and forecasting

*   Central portal acting as a single source of truth for visualising Key Performance Indicators (KPIs)

*   Enable faster decision-making to determine synergies and cost reductions as key benefits

### **TECHNOLOGIES USED**

Azure Data Factory, SAP (Ariba, BAPI, Hyperion), SharePoint, Power BI, Power Automate, PowerApps, Azure Synapse, MS-SQL

### **SOLUTION OUTCOMES**

*   Near real-time reporting and visualisation of KPIs for accelerated decision-making

*   Automation of pre-trigger activity for critical thresholds minimised human-intervention

*   Reduction in demurrage charges as a result of timely updates and data visibility to business stakeholders

*   Appreciation of technology and the realised benefits by business users made them evaluate more potential use cases of integrating technology into their day-to-day operations. Established process of enabling business users to adopt self-service BI for their reporting needs.



Smart Analytical Dashboard

The client was one of India's largest and premier banking institution providing a wide-range of financial products and services to its customers using pan-India distributed channels.

### **VISION**

As part of its digitisation journey, vendor wanted to integrate ML and AI-based approaches to redefine banking experience for its customers.

### **OBJECTIVES**

*   Minimize human intervention by automating bank statements processing using intelligent OCR solutions and achieve a high degree of accuracy.



*   Perform near-accurate account reconciliations to verify account holders and audit financial transactions.

### **CHALLENGES**

*   Different formats of bank statements for different banks

*   Manual data extraction and processing of statements

*   High turn-around time and man-hours invested for data validation

*   Chances of fraud and margin for human errors

### **TECHNOLOGIES USED**

Python, Azure Form Recognizer, Azure Custom Vision

### **SOLUTION OUTCOMES**

*   Cloud-based and real-time bulk-processing of statements

*   Automation using ML-based solutions reduced probability of fraud and margin for errors

*   Reduced turn-around time and accelerated decision-making

### **FUTURE SCOPE**

*   Scale approach to verify other KYC and user documents such as Driving License, PAN, salary slips etc.

*   Store and intelligently retrieve data for audit purposes

*   Recognise spending habits of consumers

*   Perform cash-flow analysis to offer financial services and send product recommendations

Bank Statements Automation

An in-house project focused on the use case of stock price forecasting



### **OBJECTIVES**

*   Develop an end-to-end solution for prediction of stock price using various forecasting algorithms.

*   Prepare an extensive technical and deployment framework that can be used as a ready reference for building similar solutions and for presenting capability demos to other potential vendors

*   Explore and implement various concepts including automation, dashboarding, deployment and monitoring



### **TECHNOLOGIES USED**

Databricks, MLFlow, Azure DevOps



### **SOLUTION APPROACH**

*   Scraped data of a stock for past 1 year from Money Control API

*   Data cleaning in bronze layer of Databricks

*   Stored pre-processed data in silver layer and created a feature store.

*   Using feature store, created different SOTA forecasting models such as TFT, N-beats, Deep-AR.

*   Used MLFlow for model development, experiment tracking, model management and deployment.

*   Compared multiple metrics such as RMSE, MAE, R2 Score etc. to select best performing model

*   Stored prediction in delta table and created dashboard to visualize KPIs.



Stock Price Prediction

Featured Work

View details about this [work](https://www.notion.so/Adventure-Works-Report-5227448c5d0b4afca241195722bc4e20).

*(You will be redirected to an external website)*


Power BI: Adventure Works Report

View work snapshots below. 

![](/images/Android%20Large%20-%206.png)![](/images/Android%20Large%20-%205.png)![](/images/Android%20Large%20-%204.png)


Featured Work

Smart Analytical Dashboard

Bank Statements Automation

Stock Price Prediction

Personal Projects

Power BI: Adventure Works Report

Figma: Movie Reservations