Production uptime
50M+ users
Senior .NET Backend Engineer · Fintech & Payments
I build and scale fintech and enterprise platforms that handle millions of users.
Backend for 50M+ users · 99.9% uptime · API 110s → 9–13s
Performance work is only useful when the system, team, and customers feel the difference.
50M+ users
<2h cutover
110s → 9–13s
Automated pipelines
Jenkins CI/CD
Trace the work from client request to useful response. Hover or tap a node for the optimization lens.
Clear contracts, predictable payloads, and fast feedback at the edge.
The through-line is backend ownership: understand the constraint, measure the path, ship the smallest durable fix.
Dhaka, Bangladesh (Hybrid) | Sports membership & club-management platform | .NET 9, SQL Server, Dapper
Dhaka, Bangladesh | Mission-critical .NET Core applications for 50M+ users
Dhaka, Bangladesh | Internal enterprise tools
Remote | Contract backend work
Problem, approach, result. No theatre—just the engineering decisions that moved the system.
Nationwide sales platform needed to handle 3x peak transaction volume on existing infrastructure.
Index and query-plan optimization across high-traffic paths.
Handles 3x peak volume on same infrastructure. Replaced 40-hour monthly manual reconciliation with a 2-hour automated compliance report (~450 person-hours saved/year).
Enterprise needed a modular platform serving web + mobile clients across multiple domains and regions.
Modular REST API covering membership, organization, booking, venue, invoice, class, and results domains. Multi-region Adyen payment integration with cart/store-based merchant account resolution.
Enterprise-scale platform serving web and mobile clients with multi-region payment support across UK, US, Europe, and Australia. Backend engineering contributions: Adyen integration (ACH, BACS, SEPA, PayTo), query optimization, and audit logging.
Schools need a multi-tenant management platform with strict data isolation and reliable payment reconciliation.
Built a modular monolith with Clean Architecture (Domain, Application, Infrastructure, API layers). Multi-tenant data isolation on shared infrastructure, Redis distributed caching, RabbitMQ async processing, and a payment CSV export with cent-level reconciliation logic.
Multi-tenant school management platform with tenant-level isolation, RBAC (super admin/admin/school-level) with JWT authentication, Redis + RabbitMQ for high-volume operations, and cent-level payment accuracy. Private codebase; architecture walkthrough available on request.
Teams needed a consistent way to find and review slow SQL queries and optimize them.
Collects slow-query data from SQL Server Query Store, analyzes execution plans and T-SQL, and produces copy-paste-ready suggestions, with a Blazor UI and a CLI.
Read-only by design: it never applies fixes automatically on production. Private codebase; architecture walkthrough available on request.
No progress bars. Context, constraints, and the right abstraction beat a decorative percentage.
Royal University of Dhaka
2015 - 2019 · CGPA 3.70 / 4.00 (Top 10% of cohort)
Notes on .NET, architecture and backend engineering.
Typically replies within 24h. LetΓÇÖs talk about the constraint before the solution.
Immediate to 4 weeks ┬╖ UTC+6 · Overlaps EU, US, Australia