Why do recruiters need a tech glossary?

Because titles lie and keywords do not. Knowing that a "platform engineer" lives in the DevOps family, or that PyTorch signals ML while React signals frontend, is the difference between a calibrated search and noise.

What is the difference between data engineering and data science?

Data engineers build the pipelines that move and clean data; data scientists analyze it and build models on top. Different tools, different titles, and rarely interchangeable candidates.

How current is this glossary?

It is maintained by a working technical sourcer and updated as the market's vocabulary moves; each domain lists the titles and tools seen in live searches.

The Recruiter's Tech Glossary: 12 Domains in Plain English

07 — Technology Guide

Speak the language

You don't need to code. You do need to know what these technologies are, why they matter, and how they relate to each other. This is your cheat sheet.

🐍

Programming Languages

▼

Python

The dominant language for data science, ML, AI, and automation. If you're filling any data/ML role, Python is almost always required. Think of it as the English of programming for data work.

SQL

The language for querying databases. Every data role uses SQL. It's not a programming language in the traditional sense. It's how you ask databases questions and get answers back.

Statistical programming language popular in academia, biotech, and traditional analytics. Being replaced by Python in most tech companies, but still dominant in pharma and academic research.

Java

Enterprise workhorse. Big banks, large companies, Android apps. Verbose but reliable. If a candidate has Java, they can likely pick up other languages. Strong signal for backend engineering roles.

Scala

Runs on the Java platform but more concise. Popular in data engineering because Apache Spark (the big data processing engine) was written in Scala. Scala + Spark is a classic data engineering combo.

Go (Golang)

Built by Google. Fast, simple, great for infrastructure and backend services. Popular at companies building cloud tools, DevOps platforms, and high-performance systems.

Rust

The "hot" systems language. Memory-safe without garbage collection. Growing fast in infrastructure, WebAssembly, and performance-critical applications. Rust engineers are scarce and expensive.

TypeScript

JavaScript with type safety. Standard for modern frontend and full-stack development. If a JD says "React + TypeScript," they want a frontend or full-stack engineer.

🤖

ML & AI Frameworks

▼

PyTorch

The leading ML framework. Built by Meta's AI Research lab. Dominates in research and increasingly in production. If a candidate knows PyTorch, they do real ML work, not just calling APIs.

TensorFlow

Google's ML framework. Was #1, now #2 behind PyTorch. Still huge in production deployments, especially at Google-ecosystem companies. TensorFlow experience is adjacent to PyTorch.

JAX

Google's newer ML framework. Faster and more flexible than TensorFlow. Used by cutting-edge research teams (Google DeepMind, Anthropic). JAX on a resume signals advanced ML work.

scikit-learn

The standard library for "classical" ML (not deep learning). Random forests, SVMs, clustering. Often the first ML tool people learn. Good for data scientists doing tabular data work.

Hugging Face

The GitHub of ML models. Hosts pre-trained models and the Transformers library. If someone uses Hugging Face, they work with large language models, NLP, or computer vision.

XGBoost / LightGBM

Gradient boosting libraries. The go-to for structured/tabular data problems (fraud detection, recommendations, pricing). Wins most Kaggle competitions on tabular data.

🧠

LLM & GenAI Stack

▼

LangChain

The most popular framework for building applications with LLMs. Handles chains, agents, memory, and tool use. LangChain on a resume means they build AI applications, not just use ChatGPT.

RAG

Retrieval-Augmented Generation. The technique of feeding an LLM your own documents so it can answer questions about them. The hottest pattern in enterprise AI right now.

Fine-tuning

Customizing a pre-trained model on your specific data. More advanced than RAG. Requires ML engineering skills, GPU infrastructure, and training data pipelines.

Vector Databases

Databases that store embeddings (numerical representations of text). Pinecone, Weaviate, ChromaDB, Milvus. Essential infrastructure for RAG and semantic search.

Prompt Engineering

The art of writing effective instructions for LLMs. Not a traditional engineering skill. Some companies hire dedicated prompt engineers; others expect all engineers to have this skill.

AI Agents

LLMs that can use tools, browse the web, write code, and take actions autonomously. The frontier of GenAI. Companies building agents: Anthropic, OpenAI, Google, plus hundreds of startups.

🏰

Data Engineering

▼

Apache Spark

The dominant big data processing engine. Handles datasets too large for a single machine. PySpark (Spark in Python) is the most common flavor. Core skill for data engineers.

Snowflake

Cloud data warehouse. Stores and queries massive datasets. Competing with Databricks and BigQuery. "Snowflake" on a JD means the company uses a modern data stack.

Databricks

Unified analytics platform built on Spark. Data engineering + data science + ML in one platform. Founded by the creators of Spark. Big competitor to Snowflake.

dbt

Data Build Tool. Transforms data inside the warehouse using SQL. The hottest tool in analytics engineering. "dbt" on a resume signals modern data stack fluency.

Airflow

Workflow orchestrator. Schedules and monitors data pipelines. Built by Airbnb, now Apache project. The standard for "making sure data pipelines run on time."

Kafka

Real-time data streaming platform. Handles millions of events per second. If a company uses Kafka, they process data in real-time (fintech, ad-tech, ride-sharing).

☁️

Cloud & Infrastructure

▼

AWS

Amazon Web Services. The #1 cloud platform (~32% market share). Services like S3 (storage), EC2 (compute), SageMaker (ML). AWS experience is the most in-demand cloud skill.

GCP

Google Cloud Platform. #3 cloud but dominant in data/ML (BigQuery, Vertex AI). Companies using Google's AI ecosystem often require GCP. Strong in analytics and ML workloads.

Azure

Microsoft's cloud. #2 overall, dominant in enterprise/government. Deep integration with Office 365 and Active Directory. If a company is "Microsoft shop," they're on Azure.

Docker

Containerization. Packages applications so they run the same everywhere. Nearly universal in modern engineering. If someone doesn't know Docker in 2026, that's a yellow flag.

Kubernetes (K8s)

Orchestrates Docker containers at scale. Complex but powerful. "K8s" experience means they've managed production systems. DevOps/Platform engineering staple.

Terraform

Infrastructure as Code. Defines cloud resources in config files instead of clicking through AWS console. Standard for DevOps and platform engineering roles.

⚙️

MLOps & Model Deployment

▼

MLflow

Open-source platform for ML lifecycle management: experiment tracking, model registry, deployment. The most widely adopted MLOps tool. Built by Databricks.

Weights & Biases

Experiment tracking and model monitoring. Popular in research labs and ML teams. W&B on a resume signals someone who tracks experiments rigorously, not just "vibes-based" ML.

SageMaker

AWS's ML platform. End-to-end: data labeling, training, deployment, monitoring. If a company is on AWS and does ML, they probably use SageMaker.

Feature Stores

Centralized repositories for ML features (Feast, Tecton, Hopsworks). Having "feature store" on a resume means they've built production ML systems, not just notebooks.

Model Monitoring

Tracking model performance in production. Detecting drift, bias, and degradation. Tools: Evidently, Arize, WhyLabs. Signals mature ML operations.

Speak the language

Speak the language

Programming Languages

ML & AI Frameworks

LLM & GenAI Stack

Data Engineering

Cloud & Infrastructure

MLOps & Model Deployment

Common questions

Why do recruiters need a tech glossary?

What is the difference between data engineering and data science?

How current is this glossary?

SourcingNav runs this playbook for you