Aug 2025 – Nov 2025 · 3 mos · Gurugram, Haryana, India · On-site

Nov 2024 – Feb 2025 · 3 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Migrated the entire Checkout Module from RoR to Python, integrating full payment capabilities, promo codes, and public access for self-service global shipping.
• 🛠️ Technologies Used
• Backend: Python, SQLAlchemy
• Database: PostgreSQL, Redis
• Messaging & Async Processing: Pulsar, Flower
• 🚀 Key Contributions & Features
• 1️⃣ Migrated Checkout from Ruby on Rails to Python
• ✅ Optimized database design using SQLAlchemy & PostgreSQL, improving query performance.
• ✅ Refactored legacy code, making the system easier to maintain and scale.
• 2️⃣ Full Payment Integration at Checkout
• ✅ Added direct payment processing, allowing users to pay at checkout instead of after booking.
• ✅ Integrated multiple payment gateways for seamless transactions.
• 3️⃣ Promo Code & Discount Functionality
• ✅ Implemented a promo code system, allowing users to apply discounts at checkout.
• ✅ Enabled time-limited offers, improving customer engagement.
• 4️⃣ Agent Wallet System
• ✅ Introduced an agent wallet, allowing authorized agents to use preloaded balances for transactions.
• ✅ Enabled commission-based transactions, improving agent incentives.
• 5️⃣ Opened Checkout for Public Use
• ✅ Previously, only Cogoport agents could process shipments.
• ✅ Now, anyone from anywhere can book shipments independently, enabling self-service global shipping.
• ⚡ Challenges & Solutions
• 🔹 Slow Processing Due to Bad Database Design
• 🔹 No Direct Payments at Checkout
• Solution: Integrated payment gateways, allowing real-time payments.
• 🔹 Limited User Access (Agents Only)
• Solution: Opened checkout to the public, enabling self-service global shipping.
• 📈 Business Impact
• ✅ Faster checkout processing with optimized DB queries.
• ✅ Enabled self-service global shipping, eliminating the need for agents.
• ✅ Expanded business reach, allowing any user worldwide to book shipments.
• ✅ Improved revenue through promo codes and agent wallet transactions.

Apr 2024 – Aug 2025 · 1 yr 4 mos · Gurugram, Haryana, India · On-site

Mar 2024 – Oct 2024 · 7 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• we developed an automated Supplier Payment Module that:
• ✅ Eliminated manual file handling through direct H2H (Host-to-Host) bank integration.
• ✅ Automated payment approvals through an incident-based workflow.
• ✅ Reduced processing time from 1-2 days to just 45 minutes - 1.5 hours.
• 🚀 Key Contributions & Features
• 1️⃣ H2H (Host-to-Host) Payment Integration
• ✅ Developed H2H (Host-to-Host) integration with Citi Bank, enabling direct, automated bank transactions.
• ✅ Eliminated the need for manual CSV uploads, making payments instant and error-free.
• ✅ Ensured compliance with bank security protocols using SSH-based secure file transfers.
• 2️⃣ Automated Payment Processing & Acknowledgment
• ✅ Ensured 100% accuracy in supplier settlements by eliminating manual reconciliation errors.
• 3️⃣ Error Handling & Incident Management
• ✅ If a payment failed due to incorrect bank details, the system:
• Sent alerts via email & platform notifications to stakeholders.
• Flagged the incorrect payment in the Incident Management Module.
• Allowed users to correct bank details & retry the transaction, preventing delays.
• 4️⃣ Pre-Payment Validation via ₹1 Test Transactions
• ✅ To prevent payment failures, the system first sent ₹1 test payments to verify supplier bank details.
• ✅ Reduced payment failures to just 0.2%, ensuring smooth processing.
• ⚡ Challenges & Solutions
• 🔹 Manual CSV Uploads Were Slow & Error-Prone
• 🔹 Reconciliation Delays Due to Manual Processing
• 🔹 High Payment Failures Due to Incorrect Bank Details
• 🔹 Supplier Trust Issues Due to Late Payments
• 📈 Business Impact
• ✅ Reduced supplier payment processing time from 1-2 days to just 45 minutes - 1.5 hours.
• ✅ Improved first-attempt payment success rate to 99.98%, reducing failures to just 0.2%.
• ✅ Reduced operational costs by 90% by eliminating manual processing.
• ✅ Enabled real-time tracking of supplier settlements, improving cash flow.
• ✅ Faster supplier onboarding, allowing payments & invoicing within 2-3 hours.

Sep 2023 – Mar 2024 · 6 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Before this module was built, receiving payments from customers was a highly manual process, causing:
• Long reconciliation times—It took 2-3 working days to verify payments and update invoices.
• High payment failure rates due to incorrect customer bank details.
• To automate payment collection and reconciliation, we developed a fully integrated Customer Payment Module, allowing customers to pay seamlessly via multiple payment methods while ensuring real-time tracking and reconciliation.
• 🚀 Key Contributions & Features
• 1️⃣ Payment Gateway Integration
• ✅ Integrated BillDesk, Stripe, and Razorpay to allow customers to pay via payment links.
• ✅ Enabled credit/debit card, UPI, and net banking payments, ensuring maximum flexibility.
• 2️⃣ Virtual Account System for Direct Bank Transfers
• ✅ Integrated with Axis, RBL, and IndusInd Banks to generate virtual bank accounts for customers.
• ✅ Allowed direct bank-to-bank transfers, eliminating the need for manual payment tracking.
• 3️⃣ Automated Payment Reconciliation
• ✅ Earlier, reconciliation required manual statement downloads, taking 2-3 working days.
• ✅ Now, payments are automatically matched to invoices within 1 hour.
• ✅ Used balance fetch & account statement APIs to validate payments in real time.
• 4️⃣ Error Handling & Incident Management
• ✅ If a payment failed due to incorrect bank details, the system: Sent an alert to stakeholders.
• ⚡ Challenges
• 🔹 Payments Took Too Long to Reconcile (2-3 Days Delay)
• 🔹 High Payment Failure Rates Due to Incorrect Bank Details
• 🔹 Lack of Real-Time Tracking for Payments
• 📈 Impact
• ✅ 99.98% successful first-attempt payment rate, reducing failures to just 0.2%.
• ✅ Reduced reconciliation time from 2-3 working days to just 1 hour, eliminating manual effort.
• ✅ Faster settlements improved cash flow and financial visibility.
• ✅ Automated payment tracking, reducing errors and ensuring seamless operations.

May 2023 – Sep 2023 · 4 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Before this module, overhead expenses were tracked using Excel sheets, causing:
• Lack of visibiity across different departments and countries.
• Recurring expenses manually entered every month, leading to inefficiencies.
• 🚀 Key Contributions & Features
• 1️⃣ Real-Time Expense Tracking & Categorization
• ✅ Built a centralized system to monitor expenses by department, location, and category.
• ✅ Supported multi-country operations, allowing global expense tracking.
• 2️⃣ Automated Vendor Onboarding
• ✅ Vendors could register in 5-7 minutes by submitting business details, bank info, and expense category.
• ✅ Eliminated manual approvals by automating verification.
• 3️⃣ Expense Approval Workflow via Incident Management System
• ✅ Integrated with Incident Management for automated approvals based on expense type and amount.
• ✅ Allowed real-time tracking of approvals to prevent delays.
• 4️⃣ Payrun Integration for Vendor Payments
• ✅ Seamlessly processed approved expenses via the Payrun Module, eliminating manual intervention.
• ✅ Ensured timely vendor payments, improving relationships.
• 5️⃣ Recurring Expense Automation
• ✅ Auto-generated expenses for fixed costs (e.g., rent, utilities), removing the need for manual entry.
• ✅ Sent automatic approval requests when a billing period ended.
• ⚡ Challenges & Solutions
• 🔹 Lack of Visibility Across Countries → Created a real-time dashboard for centralized tracking.
• 🔹 Slow Vendor Onboarding → Automated onboarding, reducing time from days to 5-7 minutes.
• 🔹 Manual Approval Delays → Integrated Incident Management for workflow automation.
• 🔹 Recurring Expenses Entered Manually → Automated expense generation & approval routing.
• 📈 Business Impact
• ✅ Reduced overhead expenses by 25% by identifying & cutting unnecessary spending.
• ✅ Reduced vendor onboarding time from days to 5-7 minutes.
• ✅ Improved approval speed, ensuring timely expense processing.
• ✅ Enabled real-time tracking of company expenses, improving financial decision-making.

Apr 2023 – May 2023 · 1 mo · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Before this module, approvals for financial & operational workflows were done manually, leading to:
• Delays in processing bank details, tax approvals, and financial transactions.
• Lack of transparency in approvals.
• Errors due to manual tracking.
• To address this, we built an Incident Management System to automate approval workflows for multiple financial operations.
• 🛠️ Technologies Used
• Backend: Kotlin, Micronaut
• Database: PostgreSQL, Redis
• Messaging & Async Processing: RabbitMQ
• 🚀 Key Contributions & Features
• ✅ Approval Management System for Financial Workflows
• Created an approval system for:
• Bank detail approvals.
• TDS (Tax Deducted at Source) approvals.
• Credit & Debit Note issuance.
• Payrun & Payroll approvals.
• Recurring expense approvals.
• Tax preference changes.
• ✅ Automated Incident Workflow Processing
• Used RabbitMQ to queue approval requests asynchronously, preventing system slowdowns.
• Provided real-time tracking of approval status, ensuring transparency.
• ✅ Integration with Financial Modules
• Integrated directly with invoicing, taxation, and payment systems, ensuring seamless processing of approved requests.
• ⚡ Challenges & Solutions
• 🔹 Delays Due to Manual Approvals
• Solution: Built a workflow automation engine that ensured faster approvals with predefined rules.
• 🔹 Lack of Transparency in Approval Status
• Solution: Implemented real-time approval tracking, reducing uncertainty in financial operations.
• 🔹 Errors Due to Manual Tracking
• Solution: Integrated Incident Management with the finance system, ensuring automatic data validation before approvals.
• 📈 Impact
• ✅ Improved approval efficiency, reducing delays in financial workflows.
• ✅ Ensured transparency, making it easier for teams to track approval progress.
• ✅ Reduced errors, ensuring financial compliance.

Feb 2023 – Mar 2023 · 1 mo · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Governments in India and China mandated that businesses must report invoices electronically to ensure tax compliance. Before this integration, invoice declaration was manual, leading to:
• Delays in invoice approvals & processing.
• High risk of fraudulent invoices.
• Manual errors in tax reporting.
• To comply with India’s IRN (Invoice Reference Number) requirements and China’s invoice declaration laws, Cogoport needed to integrate with:
• ✅ EY APIs (India) for IRN generation & supplier bill validation.
• ✅ Aisino APIs (China) for real-time invoice declaration to the Chinese tax authorities.
• 🛠️ Technologies Used
• Backend: Kotlin, Micronaut
• Database: PostgreSQL, OpenSearch
• Messaging & Async Processing: RabbitMQ
• 🚀 Key Contributions & Features
• ✅ India – IRN Generation & Supplier Bill Validation via EY Integration
• Integrated with EY Digisgst APIs to generate IRN (Invoice Reference Number) for invoices, ensuring compliance with India’s e-invoicing regulations.
• Integrated EY AIM module for supplier bill validation using QR code scanning, reducing invoice fraud.
• ✅ China – E-Invoice Declaration via Aisino Integration
• Integrated with Aisino Efapiao APIs to submit invoices directly to the Chinese tax authorities, ensuring compliance with government tax regulations.
• ✅ Asynchronous Processing & Performance Optimization
• Integrated OpenSearch for real-time invoice search & retrieval, improving visibility into invoicing history.
• ⚡ Challenges & Solutions
• 🔹 Handling High Invoice Volume Efficiently
• 🔹 Fraudulent Invoice Risks
• 🔹 Different E-Invoicing Regulations in India & China
• 📈 Impact
• ✅ Reduced fraudulent invoices by 50% through automated validation.
• ✅ Ensured compliance with India’s IRN requirements & China’s e-invoicing mandates.
• ✅ Automated invoice declaration, reducing manual work and errors.
• ✅ Faster invoice retrieval, improving financial reporting efficiency.

Nov 2022 – Jan 2023 · 2 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Earlier, the invoicing system was not scalable and was limited to India. Each country had different invoice formats, requiring manual invoice customization. The goal was to build a unified invoicing system that could dynamically generate invoices based on:
• Country-specific tax rules.
• Invoice structure requirements.
• Product and service category.
• The old invoicing system also had bad database design, leading to:
• Slow invoice generation times.
• Scalability issues when adding new countries.
• No support for e-invoicing compliance.
• 🚀 Key Contributions & Features
• ✅ Factory Pattern for Country-Specific Invoice Generation
• Implemented a factory-pattern-based invoicing system, allowing the platform to generate invoices dynamically based on country-specific requirements.
• Supported invoicing for India, Singapore, China, Vietnam, Thailand, Indonesia, and Hong Kong.
• ✅ Asynchronous Invoice Processing with RabbitMQ
• Instead of processing invoices synchronously, used RabbitMQ to handle invoicing asynchronously, improving system performance.
• ✅ Multi-Currency & Multi-Tax Support
• Ensured invoices were generated in the correct currency and included country-specific tax details.
• ✅ Improved Database Design for Scalability
• Optimized database queries and restructured invoice storage, reducing invoice generation time significantly.
• ⚡ Challenges & Solutions
• 🔹 Different Invoice Formats Across Countries
• Solution: Used a factory pattern to dynamically generate invoices based on country-specific rules.
• 🔹 Slow Invoice Processing Due to Bad DB Design
• Solution: Redesigned DB schema for optimized queries and improved caching for frequently accessed invoice templates.
• 🔹 No Support for Real-Time Invoice Processing
• 📈 Impact
• ✅ Standardized invoicing across 7 countries, eliminating the need for manual invoice customizations.
• ✅ Reduced invoice generation time, improving overall efficiency.
• ✅ Prepared the system for future e-invoicing compliance, making tax reporting easier

Jun 2022 – Oct 2022 · 4 mos · Gurugram, Haryana, India · On-site

• 📌 Purpose
• Cogoport, a digital freight logistics platform, initially developed a taxation module only for India. This system computed tax based on some factor
• Service type (e.g., freight, customs, warehousing).
• Trade type (e.g., SEZ exemptions, GST mechanisms).
• HS codes (Harmonized System codes for product classification).
• As Cogoport expanded internationally, this system faced severe limitations:
• 🔹 No Global Scalability: It couldn’t adapt to multi-country taxation.
• 🔹 Performance Issues: Poor database design slowed computations.
• 🔹 High Response Times: Synchronous tax processing led to delays.
• To solve these challenges, we built a global taxation module that:
• ✅ Supported multiple countries and dynamically applied tax rules.
• ✅ Improved performance using Redis caching.
• ✅ Reduced response time with asynchronous processing via RabbitMQ.
• 🛠️ Technologies Used
• Backend: Kotlin, Micronaut
• Database: PostgreSQL, Redis
• Messaging & Async Processing: RabbitMQ
• 🚀 Key Contributions & Features
• The new taxation module supported seven countries:
• 🌍 India, China, Thailand, Vietnam, Indonesia, Singapore, and Hong Kong.
• Each country had different tax regulations, requiring dynamic tax calculations based on:
• Country-specific VAT/GST rates.
• Product classification (HS codes).
• Special exemptions like SEZ in India.
• Result: Global scalability & compliance without extra manual effort
• ✅ Redis Caching for Performance Optimization
• 🔹 Problem: Every tax computation queried the database, leading to high latency.
• 🔹 Solution: Redis caching stored frequently used tax rules.
• Results:
• ✅ Reduced database load by 70%.
• ✅ Cut API response times by 60-80%.
• ✅ Faster tax computations with minimal delay.
• 📈 Impact & Results
• 🚀 Enabled tax support in 7 countries, allowing global expansion.
• 🚀 Optimized system performance, reducing response time by 60-80%.
• 🚀 Reduced manual tax handling, ensuring seamless compliance.
• 🚀 Automated tax calculations, improving financial efficiency.

Aug 2018 – Jul 2022 · 3 yrs 11 mos · Mizoram, India