Medium

Pravash

In

Level Up Coding

by

Pravash

·

Jul 1

Memorizing Design Patterns Is a Waste of Time. Use This Decision Tree Instead.

How to go from “I know what Strategy is” to “I know when to use it”

This technical infographic visualizes how to map complex software engineering bottlenecks to standard design patterns. It flows left-to-right from a core structural question mark to clear, architectural outcomes.

In

Level Up Coding

by

Pravash

·

Jun 15

Delta Lake vs. Iceberg vs. Hudi: I Finally Know When to Use Each One

Three open table formats. Three different jobs. One cheat sheet.

Delta Lake vs Apache Iceberg vs Apache Hudi — open table format comparison for data engineers

In

Level Up Coding

by

Pravash

·

Jun 1

LLM vs RAG vs MCP: I Finally Know When to Use Each One

Three architectures. Three different jobs. One cheat sheet.

LLM vs RAG vs MCP: I Finally Know When to Use Each One

In

Level Up Coding

by

Pravash

·

May 18

8 Data Pipeline Patterns Exist. Most Engineers Only Know 3

Most pipelines fail not because of bad code — but because engineers picked the wrong pattern from the start.

The image conveys strategic planning and system design, visually representing the complexity and structure of modern data pipelines and engineering workflows.

In

Level Up Coding

by

Pravash

·

Apr 28

The 5 Creational Pattern — I finally understand when to use each one

Master the 5 Creational Design Patterns in Python with real-world stories even a 10-year-old can follow

Five Python creational design patterns listed on a dark and light split layout.

In

Level Up Coding

by

Pravash

·

Apr 20

PyResilience: One Decorator. Seven Patterns. Zero Excuses.

Tenacity, pybreaker, hand-rolled timeouts. They don’t talk to each other. That gap is your next incident.

Pyresilience Python resilience graphic with retry, cache, fallback, rate limiting, circuit breaker, timeout, and bulkhead.

In

Level Up Coding

by

Pravash

·

Apr 13

Composition over Inheritance in Python (Explained With a Real Refactor)

A practical refactor pattern you can reuse to make Python systems easier to test, extend, and debug.

Inheritance vs Composition diagram in OOP showing monolithic class hierarchy vs modular, swappable components for flexible system design

In

Level Up Coding

by

Pravash

·

Apr 7

Stop Using PySpark UDFs Like This — Here’s the Faster Pattern

The fastest way to cut UDF-heavy stage time without rewriting your pipeline in Scala.

PySpark UDF performance: slow row-based vs fast vectorized Spark with Apache Arrow optimization

In

Level Up Coding

by

Pravash

·

Mar 23

Databricks Auto Loader: 7 Production Failures I Wish Someone Warned Me About

Checklist for schema drift, CloudFiles gotchas, and streaming pipelines that fail

Schema drift breaking data pipeline: Bronze, Silver, Gold layers with ADLS, Auto Loader, and streaming data issues

In

Level Up Coding

by

Pravash

·

Mar 4

5 Python Libraries That Stop OOMs in Memory-Bound ETL

Process TB-scale data with drastically lower RAM inside Prefect pipelines (with real-world tradeoffs)

ETL pipeline optimization diagram showing memory-bound transforms, data flow, and performance monitoring system

Pravash

Pravash

I am a passionate Data Engineer and Technology Enthusiast. Here I am using this platform to share my knowledge and experience on tech stacks.

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech