JS .NET
Data Engineer · Kathmandu, Nepal

Prashanta Babu Thapa Building pipelines that matter.

Associate Data Engineer at dlytica — designing scalable ELT pipelines for Nepal's banking sector. Airflow-orchestrated medallion architectures using Apache Spark, dbt, Trino, Iceberg, and NiFi. Previously a Full-Stack Developer.

See Architecture Get in Touch GitHub ↗ LinkedIn ↗
1.4+Years Data Eng
10moFull Stack Dev
12+Tools mastered
Medallion Architecture

Airflow-Orchestrated
Data Pipeline

Airflow orchestrates the entire pipeline — schedules NiFi to extract from Oracle, MongoDB, MySQL & MSSQL into Bronze Iceberg, then triggers Spark (Bronze→Silver) and dbt (Silver→Gold). Trino queries the Gold layer. Superset renders the dashboards.

Live simulation · Medallion Architecture with Airflow as the central orchestrator

Tech Stack

Tools &
Technologies

Modern data engineering toolkit centered on Airflow orchestration and lakehouse architecture.

Orchestration — Heart of the Pipeline
Apache Airflow
Airflow
DAG SchedulingPipeline OrchestrationPython Operators
Ingestion Layer — Bronze
NiFi + Source Systems
NiFi
Oracle
MongoDB
MySQL
MSSQL
Apache NiFiOracleMongoDBMySQLMSSQL
Processing — Silver Layer
Apache Spark
Spark
PySparkSpark SQLDelta/IcebergStreaming
Transformation — Gold Layer
dbt
dbt
Trino
Spark
dbt-sparkdbt-trinoLineageTests
Storage — Iceberg Lakehouse
Iceberg + MinIO
Iceberg
MinIO
Apache IcebergMinIO / RustFSParquet
Query & Analytics
Trino + Superset
Trino
Superset
Trino SQLApache SupersetDashboards
Infrastructure
Platform Ops
KubernetesDockerHelm
Web Development
Full Stack
React.jsNode.js.NETPython
Career Path

Work
Experience

From building web interfaces to orchestrating production-scale medallion data pipelines.

Jan 2025 — Present · 1 yr 4 mos
Associate Data Engineer
dlytica Private Limited · Data Analytics & AI · Full-time · Kathmandu
Building the AI360 platform (Customer 360 + Branch 360) for NabilBank using Airflow-orchestrated medallion architecture. NiFi ingests from Oracle CBS and other sources into Bronze (Iceberg/MinIO). Spark refines into Silver. dbt transforms into Gold. Trino queries Gold for Superset dashboards. Kubernetes for all infra. Analytics cover NFRS 9 ECL, Loan Loss Provisioning, dormancy, and branch profitability.
Airflow
NiFi
Spark
dbt
Iceberg
Trino
Superset
Kubernetes
Jul 2021 — Apr 2022 · 10 mos
Junior Software Developer
General Technology Pvt. Ltd. · Full-time · Baluwatar, Kathmandu
Built full-stack web applications using React.js, Node.js, and .NET. Focused on scalable, user-focused interfaces with strong separation of concerns. Delivered performant, secure, and maintainable products.
React.js
Node.js
.NET
Education
Bachelor's in Engineering
Kathmandu Engineering College · Tribhuvan University
Engineering degree providing the systems thinking foundation applied in distributed data infrastructure, pipeline design, and cloud-native architecture.
📍 Kathmandu, Nepal
Selected Work

Projects &
Builds

Real-world data engineering and software problems solved in production.

01
NabilBank AI360 — Core Banking Platform
Airflow-orchestrated medallion lakehouse for Nepal's leading commercial bank. NiFi extracts from Oracle CBS, MongoDB, and other core banking sources into Bronze Iceberg layer. Apache Spark refines to Silver. dbt transforms to Gold. Trino queries for Superset dashboards. Covers Customer 360, Branch 360, NFRS 9 ECL provisioning, loan analysis, and dormancy detection.
AirflowNiFidbt IcebergTrinoSupersetK8s
View on GitHub →
02
ELT Pipeline — Airflow + dbt + Snowflake
End-to-end ELT pipeline using Apache Airflow for orchestration, dbt for data transformation, and Snowflake as the cloud data warehouse. Automates ingestion, transformation, and loading with scalable DAG scheduling. dbt models handle staging, intermediate, and mart layers with full lineage and testing.
Apache Airflowdbt SnowflakeELTPython
View on GitHub →
03
ORMS — Operational Risk Management System
Full-stack web application for Operational Risk Management built at General Technology. Focused on the Loss Data module — capturing, classifying, and reporting operational loss events for financial institutions. Built with React.js frontend, Node.js API, and .NET backend services with role-based access control.
React.jsNode.js .NETRisk Analytics
View on GitHub →
// Get In Touch

Let's Build
Something Together

Open to data engineering roles, lakehouse architecture projects, and interesting pipeline challenges.

[email protected]

+977-9810288411 · Kathmandu, Nepal

LinkedIn GitHub