Software builder · Educator

Tanjim Reza

Adjunct Lecturer
BRAC University
Software Engineer II
TheProductDev LLC

I design data platforms, backend systems, and automation for teams that need reliable software and clear decisions.

$10K+ Upwork earnings across client systems
100% Upwork job success
10h → 30m Weekly processing reduced through automation
Adjunct Lecturer BRAC University CSE Department
IEEE 2024 Published Bangla character research

Systems built for work that cannot afford to break.

Three anonymized engagements across financial data, analytics, and regulated call operations, designed around reliability, clarity, and measurable improvement.

Financial data pipeline

From scattered reports to a dependable data flow.

A secure admin platform connected financial sources, automated reporting cycles, validated incoming data, and delivered analytics-ready datasets.

Connect Extract Validate Schedule Warehouse
  • Multi-connection OAuth with refresh and revoke flows
  • Automated A/R and A/P aging reports
  • Daily and monthly job orchestration
  • Validation feedback and fault-tolerant execution
DjangoPostgreSQLOAuth 2.0ETLCloud RunBigQuery
Analytics API

One reporting layer for financial and sales decisions.

A multi-tenant API translated scheduled warehouse loads into consistent reporting for revenue, customer activity, vendor performance, aging, cohorts, and recurring revenue.

  • JWT authentication and role-based admin access
  • Ordered loaders with partial-failure handling
  • Standardized metrics and CSV exports
  • Rate limits and structured error responses
PythonFlaskPostgreSQLJWTBigQueryETL
Call analytics

Live visibility for regulated conversations.

A webhook-driven platform combined transcription, sentiment analysis, compliance checks, and operational dashboards for collections workflows.

  • Automated call ingestion and processing
  • Speech-to-text and sentiment analysis
  • Policy and regulatory compliance checks
  • Performance rankings, KPIs, and risk alerts
FlaskStreamlitPostgreSQLSpeech-to-TextLLM AnalyticsWebhooks

Capabilities

The engineering range behind each dependable system.

Data

Data platforms & ETL

Pipelines, validation, scheduled loads, warehouse integration, and reporting flows.

Software

Backend systems & APIs

Django and Flask services, authentication, multi-tenant APIs, and admin platforms.

Operations

Deployment & operations

Google Cloud deployments, IAM, databases, billing, monitoring, Docker, GitHub Actions, and production jobs.

Applied ML

Applied ML & automation

Computer vision, speech processing, LLM-assisted analytics, and workflow automation.

Teaching

Teaching & communication

Teaching data structures, operating systems, microprocessors, and databases through clear models and hands-on work.

Experience

Build it. Explain it. Improve it.

Engineering and teaching reinforce the same habit: understand a system deeply enough to make it dependable and clear enough for someone else to use.

Software Engineer II

TheProductDev LLC

Building software products, backend services, data pipelines, and automation in Python and Node.js, while also owning the GCP environments that run them.

  • Replaced an owner-dependent set of remembered deployment commands with a documented, scripted workflow, reducing the procedure from roughly 20 minutes to 2 minutes.
  • Reduced monthly GCP costs by 25% through resource right-sizing, idle-resource cleanup, scaling changes, and query optimization.
  • Used Cloud Logging, Cloud Monitoring, and uptime checks to surface Cloud Run jobs, BigQuery jobs, and QuickBooks OAuth failures for immediate resolution.
  • Integrated Vertex AI APIs into product features and automated Docker image builds with GitHub Actions.
Cloud RunCompute EngineBigQueryCloud SQLCloud StorageSecret ManagerCloud SchedulerVertex AI

Undergraduate Teaching Assistant

BRAC University

Tutored computer science students, evaluated assignments, and supported lab sessions across core programming courses.

Published research

IEEE ICEEICT 2024

Teaching machines to see inside Bangla compound characters.

“Segmentation of Bangla Compound Characters” introduced the BanglaBorno dataset and applied YOLOv8 to detect and segment the simple characters inside handwritten compounds, with VGG-16 used in the research pipeline.

YOLOv8VGG-16BanglaBornoComputer Vision
Read the publication ↗

About

I’m most useful where technical depth, practical delivery, and clear communication need to coexist.

My background combines computer science at BRAC University with hands-on engineering across data platforms, APIs, automation, machine learning, full-stack products, and IoT. Teaching keeps my fundamentals sharp; client work keeps the outcomes real.

Academic
Merit Scholarship RecipientBRAC University
Security
Top 3 CTF Team in BangladeshTeam Agrani · CTFtime
Leadership
Director, Creative DepartmentBRAC University Computer Club
Service
11 years of scouting leadershipBoys Scout & Rover Scout Leader

Start a conversation

Have a complicated system? Let’s make it clear.