NPBlue is a working engineer's reference library — 800+ guides across Python, SQL, Spark, the three major clouds, and modern AI, each written to take you past the tutorial and into production.
How NPBlue works
Most sites stop at the article. Here, every topic connects to something you can actually do — hands-on challenges, error lookups, and interview prep built around the same material.
Structured tracks that build concept on concept — no orphaned one-off posts, no assumed magic.
Pick a track →Write real Python, SQL, and Spark SQL against hidden test cases — right in your browser, nothing to install.
Start solving →Paste any stack trace and get ranked matches from 800+ articles that explain or fix it.
Find a fix →Interview question banks and behavioral frameworks that turn what you've learned into offers.
Prep now →Core languages
Three foundations that carry every data and backend role — covered from first principles through the edge cases that only show up in production.
The free toolbox
Every tool below works in your browser, costs nothing, and needs no signup. This is what makes NPBlue different from a blog.
Python, SQL, and Spark SQL challenges with hidden test cases, XP, levels, and daily streaks.
Start practicing →Paste a Terraform, Spark, AWS, or Snowflake error and get ranked article matches instantly.
Paste an error →Terraform, dbt, SQL window functions, and AWS CLI — condensed to what you copy-paste mid-task.
Open a sheet →Compare token pricing across OpenAI, Anthropic, Google, and open models before you commit.
Estimate costs →Cloud cost, storage scaling, transfer time, GPU memory, salary benchmarks, and a dozen more.
Browse calculators →Curated remote and hybrid data, cloud, and AI roles — updated continuously, filtered for signal.
See open roles →Cloud platforms
Most engineers end up touching more than one cloud. These tracks use consistent depth and structure across all three, so switching platforms doesn't mean starting over.
106 guides — the deepest track on the site. Compute, storage, databases, networking, and the exam-favorite details.
Explore AWS →BigQuery, Dataflow, Vertex AI, and GKE — Google's data-first cloud, explained for engineers shipping real workloads.
Explore GCP →Synapse, Data Factory, and AKS — the platform most enterprises already run, minus the marketing gloss.
Explore Azure →Data engineering
Transformation, warehousing, and security — the three layers where data careers are actually made, covered by people who run them in production.
60 guides on models, tests, macros, and incremental strategies — version-controlled SQL done properly.
Read the track → SecurityEncryption, RBAC, zero-knowledge patterns, and compliance — the part everyone skips until an audit.
Read the track → WarehouseVirtual warehouses, zero-copy cloning, Snowpark, and the cost controls that keep finance off your back.
Read the track →Artificial intelligence
Six connected tracks that build from ML fundamentals through deep learning to the GenAI systems teams are shipping right now.
Prompting, fine-tuning, evaluation, and deployment — everything that comes after the demo works.
Explore GenAI →Tool use, memory, planning, and the Model Context Protocol — agents that survive contact with reality.
Explore Agents →Chunking, embeddings, hybrid search, and retrieval evaluation — grounding models in your own data.
Explore RAG →Supervised learning through model evaluation — the classical foundations every AI role still tests for.
Explore ML →A 50-guide curriculum from the perceptron to transformers — the math explained, never hand-waved.
Explore Deep Learning →Quantization, on-device inference, and the constraints of running models where the cloud can't reach.
Explore Edge AI →Career
The same guides you learn from, reorganized around what interviewers actually ask — technical rounds, behavioral rounds, and the salary conversation after.
