// Data Engineer · Kathmandu, Nepal · Open to Remote

PRASHANTNEPAL

Spark Optimization · Lakehouse Architecture · Kubernetes Platforms

I build the infrastructure that makes data useful at scale — terabyte pipelines, Iceberg lakehouses, and K8s platforms that actually survive production.

Let's Talk → Play the Game ↓ ↓ Download Resume

80%

Pipeline Speedup

4×

Spark Perf Gain

99%

Job Success Rate

3TB+

Daily Volume

SCROLL

// 02 · Arsenal

Stack, ranked by depth.

Expert — Daily Production

⚡ Apache Spark

☸ Kubernetes/OpenShift

🐳 Docker & Helm

🧊 Apache Iceberg

🐍 PySpark/Python

🌊 Airflow

🟠 Apache NiFi

📦 Parquet/Avro/ORC

Solid — Built & Shipped

🔷 Trino

📡 Kafka

🗄 Hive Metastore

🧪 DBT

🪣 MinIO/S3

📊 Superset

☁ Azure

🔑 Lakekeeper

Working Knowledge

⭐ StarRocks

🎯 Doris

☕ Java

⚙ CI/CD

🗃 Oracle/PostgreSQL

// 03 · Experience

Where I've shipped.

Feb 2025 — Present

DLytica Inc.
Client: Ooredoo
Kathmandu, Nepal

Data Engineer — Platform & ETL

4× Spark performance — rewrote partitioning, broadcast joins, shuffle configs. Hours → minutes at TB scale.

Spark Connect on OpenShift — centralized compute for 15+ devs, 70% faster onboarding.

Oracle → Iceberg lakehouse — bronze/silver/gold with ACID guarantees. Analytics latency cut 60%.

3TB+ daily ingestion — NiFi + Kafka from SFTP, Oracle CDC, REST APIs at 95% reliability.

85% error reduction — automated reconciliation and validation frameworks.

SparkOpenShiftIcebergTrinoAirflowNiFiKafkaDBT

Feb 2025 — Present

DLytica Academy
Kathmandu, Nepal

Tutor — Data Engineering

Mentored 10+ Data Engineering Fellows on Spark, K8s, DBT, Airflow, and Iceberg.

Feb 2023 — Jan 2024

Offer Sewa Pvt. Ltd.

Junior Software Engineer

REST APIs in Laravel/Python. 40% response time reduction via query indexing.

LaravelPythonPostgreSQL

// 04 · Impact

Numbers I own.

80%

Pipeline Runtime Cut

Terabyte Spark workloads from hours to minutes.

Spark · OpenShift

4×

Spark Performance

Broadcast joins, partitioning, executor rewrite.

Spark SQL · PySpark

99%

Job Success Rate

K8s Spark Operator with tuned pod templates.

Kubernetes · Helm

85%

Error Reduction

Automated reconciliation, zero silent corruption.

Data Quality

3×

Read Throughput

Parquet 128MB blocks, pruning, column projection.

Parquet · MinIO

70%

Faster Onboarding

Spark Connect unlocked 15 engineers overnight.

Spark Connect · DBT

// 05 · Interactive Terminal

Type `help` to explore.

prashant@data-platform:~$ spark-cluster [connected]

╔══════════════════════════════════════════════════════╗ ║ Prashant Nepal · Data Platform Terminal v2.0 ║ ╚══════════════════════════════════════════════════════╝ Type 'help' for available commands.

prashant@platform:~$

// 06 · Data Pipeline Defender

Protect your lakehouse.

Corrupt data packets are attacking your Bronze layer. Move with A/D or ← →, shoot with Space. Earn points, unlock badges, survive 5 waves. Can you reach Gold tier? 🏆

0SCORE

♦♦♦INTEGRITY

1WAVE

BRONZETIER

HOW TO PLAY

← → or A/D — move defender

SPACE — fire validation query

Red packets — corrupt data (avoid or shoot)

Purple packets — schema drift (worth 2×)

Orange packets — duplicates (shoot fast!)

BRONZE → 0 pts

SILVER → 300 pts