Back to blog
FILE 0xD0·EVERCV NOW TRACKS DATA ENGINEERING WORK: AIRFLOW DAGS, AIRBY

EverCV now tracks data engineering work: Airflow DAGs, Airbyte syncs, Dagster pipelines, Amplitude annotations, Metabase dashboards

June 21, 2026 · evercv, dataengineering, analytics, buildlog

EverCV is now at 120 signal sources.

Tonight's batch covers a gap I've been thinking about since the beginning: data engineering work. The modern data stack is one of the most active and high-leverage parts of infrastructure, and it's almost entirely invisible to every CV tool that exists.

The problem with data engineering resumes

A backend engineer opens their GitHub profile and sees 3,000 commits. That's a resume in rough form — still needs formatting, but the raw material is there.

A data engineer opens their GitHub profile and sees maybe 200 commits. Because the actual work is running: Airflow DAGs that ran 800 times this quarter, Airbyte syncs that moved 40TB of data, Dagster pipelines that backfilled 18 months of revenue metrics. The work is in the orchestration layer, not the git history. And the business impact — "I built the pipeline that powers the weekly board deck" — never shows up in a commit diff.

Same story for analytics engineers. The Metabase dashboard you built for the growth team gets used 50 times a day. The Amplitude annotation you dropped when the new onboarding flow shipped was the signal that made the A/B test legible. None of that is in git.

The five new sources

Airflow — fetch_dag_runs_for_day() pulls from the Airflow REST API v1 (/api/v1/dagRuns) using Basic auth. It returns successful runs for the day, grouped by DAG. The resume bullet reads like: "Maintained data pipeline customer_ltv_daily (Apache Airflow) — ran successfully 24 times, averaging 4m 12s per run." That's the kind of operational ownership that belongs on a CV but never gets written down.

Airbyte — fetch_syncs_for_day() hits the Airbyte Cloud API (/v1/jobs) with Bearer auth and cursor-based pagination. Filters for jobType=sync + status=succeeded. Each sync completion becomes a CV entry with source connector name and destination. If you've been keeping customer data flowing reliably from 15 sources to Snowflake for two years, that's infrastructure ownership — and now it's on your CV.

Dagster — fetch_runs_for_day() posts to Dagster Cloud's GraphQL endpoint with Dagster-Cloud-Api-Token authentication. The GraphQL query filters by endTime Unix timestamp for the day. Dagster is increasingly the orchestrator of choice for data teams that want strong typing and observable pipelines. Your pipeline runs now create CV entries.

Amplitude — fetch_annotations_for_day() pulls annotations from the Amplitude API using Basic auth (api_key:secret_key). In analytics-driven organizations, dropping an Amplitude annotation is how you mark "this feature shipped" in the metrics. It's the timestamp that makes before/after comparisons legible. EverCV now picks those up as CV signals — "shipped feature annotations in Amplitude for 4 releases Q2 2025."

Metabase — fetch_questions_for_day() authenticates via POST to /api/session then queries /api/card?f=all, filtering for cards updated today by the authenticated user. Metabase questions and dashboards are the artifacts analytics engineers spend a large part of their days building. If you created or significantly updated a dashboard, that's deliverable work — and it belongs on your CV.

Why this matters for the data engineering audience

The data engineering job market has gotten competitive in the last two years. Salaries compressed, hiring slowed, and the bar for "senior data engineer" got higher. At the same time, the tooling landscape fragmented — dbt, Prefect, Airflow, Dagster, Airbyte, dlt, Meltano — and demonstrating breadth across the modern data stack became genuinely valuable.

EverCV with the data engineering adapters means:

  1. Your orchestration ownership is captured automatically (Airflow, Dagster, Prefect — added last week)
  2. Your data movement work is captured (Airbyte)
  3. Your analytics artifacts are captured (Metabase, Amplitude)
  4. Your data transformation work is captured (dbt — also added last week)

That's the full stack, from ingest to dashboards, showing up in your CV without any manual entry.

Current source count

120 total signal sources across 60+ platforms. The data engineering row in the source table now looks like:

| Layer | Sources | |---|---| | Orchestration | Apache Airflow, Dagster, Prefect Cloud | | Ingest / ELT | Airbyte | | Transformation | dbt Cloud | | BI / Analytics | Metabase, Amplitude |

That's the complete modern data stack, minus Spark/Databricks (next).

If you're a data engineer or analytics engineer and want to see what EverCV would pull from your connected tools, free tier starts at evercv.io — no credit card, GitHub only for free, all data engineering sources on Prosumer ($15/mo).