Skip to content
Career paths · updated for 2026 hiring · Last updated 2026-05-10

Pick your data career.Then pick your first project.

Five paths, one training platform. Every path is built from the same 30+ courses and projects — the order and selection are what change. Find where you fit in 30 seconds.

Compare all 5 paths
5
distinct career paths
30+
shared courses / projects
Free
first module, every path
2026
AI-era curriculum
Start here

Who are you today?

Pick the closest match. We’ll highlight the path that usually fits — you can still explore the others.

How the paths connect

One curriculum. Five orderings.

All 5 paths share the same library of 30+ courses. Start with one — you can always pivot without starting over.

01 Foundations
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
02 Batchshared
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
03 Streaming
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
04 Quality
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
05 AI & Vectorsshared
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
06 Platform
Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer

Module 02 (Batch) and Module 05 (AI & Vectors) appear in most paths — they're the shared backbone of modern data work. Start with either and you're moving forward regardless of which role you end up in.

How to choose

5 data-engineering career paths in 2026 — pick yours.

“Data engineer” is no longer one job. In 2026 the market has split into five distinct roles with different daily work, different interview rubrics, different salary bands, and different on-call expectations. Picking the right one isn’t about which tools you know — it’s about which problem you want to solve every day for the next 3–5 years.

The 5 roles, in one sentence each

  • Core Data Engineer — Build the warehouse-ready pipelines that BI + product teams run on. SQL + Python + dbt + Airflow + Snowflake. The most common entry point.
  • AI Data Engineer — Build the RAG + LLM + agentic infrastructure that AI products run on. Vector databases, retrieval pipelines, LLM evaluation, cost-tracked serving. The fastest-growing path.
  • Data Platform Engineer — Build the lakehouse, IaC, and multi-team platform that everyone else builds on top of. Iceberg, Terraform, CI/CD, cost-attribution, RBAC. For senior+ engineers.
  • Analytics & Product Data Engineer — Build the metrics layer, experimentation infra, and KPI architecture product teams ship features on. dbt + MetricFlow + CUPED. For engineers who like working close to the business.
  • AI/ML Platform Engineer — Build the feature stores, model deployment pipelines, and inference-serving fabric ML teams ship models on. MLOps + feature stores + vLLM + drift detection. The senior-to-staff path for engineers in AI infra.

How to pick yours

Three questions answer this faster than any quiz: (1) Do you want to work with business teams or with engineering teams? Business-leaning → Analytics + Product DE. Engineering-leaning → everything else. (2) Do you want to ship features or build platforms? Features → Core DE or AI DE. Platforms → Platform DE or AI/ML Platform. (3) Where do you want to be in 5 years — staff engineer, technical lead, or director? Staff (depth) → Platform or AI/ML Platform. Lead (breadth) → Core DE or Analytics DE. Director (people) → any of the above plus explicit management track.

How AI-DE’s 5 paths are different

Most career-path content stops at “here’s the role description.” The five paths below go further: each one specifies the architectures you’ll defend in interviews, the salary band by seniority, the median time-to-first-offer for switchers, the on-call expectations, and the named hiring rubric (system design / coding / behavioral split). Every path is built from the same 32-curriculum, 30+-project core — what differs is the order, the emphasis, and the capstone you ship as your interview portfolio.

What “senior” means in each path

The senior+ bar is different per role. Core DE senior owns a domain (orders, billing, growth) and the pipelines that serve it end-to-end. AI DE senior owns a retrieval system or an LLM pipeline and the eval framework that gates its releases. Platform senior owns infrastructure that other engineers depend on — and the SLA for it. Analytics senior owns a metrics layer + experimentation system + a quarterly stakeholder review cadence. AI/ML Platform senior owns the feature store, model deployment, and the drift detection that keeps production models honest.

Below: the side-by-side comparison

The grid below shows each path’s daily-work shape, hiring signal for 2026, salary band, and capstone project. Pick the closest match — if more than one fits, optimize for the day-to-day work, not the salary band (the bands converge at senior+ across all five paths, and you’ll spend 90% of your career in the daily work).

Five career paths

Compare the five.

Each card shows what you’ll actually build, how long it realistically takes, and what the market is asking for in 2026.

The core pathpath / de

Core Data Engineer

Own Airflow DAGs, dbt models, Spark jobs, and the lakehouse they feed. Ship 6 production projects, each one interview-ready.

12–24 weeksTimeline
P04Capstone
4+Projects
SQLAirflowdbtSparkKafkaIceberg
Highest-volume DE hiring
See Core Data Engineer plan
Fastest-growing rolepath / aide

AI Data Engineer

Vector retrieval, feature stores, LLM batch enrichment, evaluation pipelines. The infra that makes AI actually work in production.

14–20 weeksTimeline
P06Capstone
4+Projects
EmbeddingspgvectorQdrantFeastRayLLM APIs
AI-era specialty · top of stack
See AI Data Engineer plan
Senior / staff trackpath / dpe

Data Platform Engineer

Multi-tenant Airflow, cost attribution, self-service tooling, SLAs. The work that gets you promoted to staff.

16–24 weeksTimeline
P12Capstone
4+Projects
KubernetesTerraformAirflowObservabilityIAMCost tooling
Senior-only hiring
See Data Platform Engineer plan
Closest to the businesspath / ae

Analytics / Product Data Engineer

Model the warehouse so analysts can trust it. dbt-first, contract-driven, semantic-layer native. Bridge data and product.

8–16 weeksTimeline
P19Capstone
4+Projects
SQLdbtSnowflakeData modelingContractsSemantic layer
Growing fast across the SaaS stack
See Analytics / Product Data Engineer plan
Senior+ AI infrapath / aip

AI Platform Engineer

Multi-tenant LLM serving, model registries, feature stores, drift monitoring. The platform layer that makes production AI actually work — and the work that gets you to staff.

20–28 weeksTimeline
P18Capstone
4+Projects
BentoMLTritonvLLMRayMLflowK8s
Senior+ only · platform ownership
See AI Platform Engineer plan
How to choose

Side-by-side, no fluff.

The honest differences. Pick based on what you actually want your day to look like.

Core Data EngineerAI Data EngineerData Platform EngineerAnalytics / Product Data EngineerAI Platform Engineer
Best if you…Want broad demandWant the AI-era edgeAlready shipping, want scaleLove SQL + businessAlready senior, want AI infra
Time to role3–6 mo3–5 moSenior only2–4 moSenior+ only
Coding neededMediumMedium–HighHighLow–MediumHigh
Math / MLNoneLowNoneNoneLow
On-callYesSometimesYesRareYes
Hiring signal · 2026🔥 Hot (high demand)🚀 Exploding (fastest hiring)Senior-only↗ GrowingSenior+ only

Already building? Skip the basics.

Jump straight to the advanced projects that stretch you — AI data infra, streaming at scale, platform engineering.

Browse advanced projects
Still not sure?

Take the 2-minute assessment.
We’ll pick your path.

Five questions. A personalized plan. Free first module unlocks in your inbox.

Just show me projects
30+ PROJECTS · 5 PATHS · ONE PLATFORM
Press Cmd+K to open