◆ Astronomy

What a data scientist astronomy
really does.

22 tasks, each one witnessed by the sources that watched the job — and behind every one, a prompt you can use tonight.

22evidenced tasks
262,440in the US (2025)
$120,230median pay / year
8systems it runs on
See the 22 tasks ↓Open in the atlasThe full career map →
This is what one task looks like here
Manage large amounts of data
Reorganise the observatory's raw and reduced data lakes: define partit…3 sources agree

The shape of the day

tap a movement to see its tasks

Which one is you, right now?

Pick the moment · no score, no sign-up
Which moment is me right now?
Whichever you pick, the task behind it opens below.

The work, task by task

22 tasks
Hands on the work18
Manage large amounts of data+
Reorganise the observatory's raw and reduced data lakes: define partitioning and retention rules, implement an ETL that deduplicates and cleans telemetry, benchmark query performance on representative 100 TB, and document the pipeline for the ops team.
escoonetwiki3 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Present findings through reports and presentations+
Draft a results report and a 12‑slide presentation summarising detection rates, model limitations, and recommended observing changes, include key plots and tables from the latest run, and circulate to the instrument scientist and head of operations by Tuesday.
escojdonet3 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Communicate insights to stakeholders+
Prepare a one‑page insights brief and a 10‑minute demo for the program manager and data curator explaining the new classifier behaviour, actionable recommendations for pipeline thresholds, and where further labeling is required; request a 30‑minute follow‑up next week.
escojdonet3 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Apply machine learning techniques+
Train and benchmark a suite of supervised and unsupervised algorithms on the cleaned lightcurve dataset, log hyperparameters and performance comparisons, and produce a reproducible pipeline so the team can rerun experiments on new runs.
jdonet2 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Merge data sources+
Combine spectroscopic, photometric, and telemetry tables into a single canonical dataset with unified timestamps and provenance, resolve schema mismatches, flag uncertain joins, and produce a versioned output the analysis team can use.
escojd2 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Combine statistical knowledge with coding+
Combine the telescope logs, calibrated photometry tables and survey metadata into a reproducible analysis pipeline, implement the statistical models and productionize the code so colleagues can rerun nightly batch recommendations.
wiki
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Visualize data to identify patterns+
Create interactive plots of light curves, color–magnitude diagrams and spatial density maps to reveal periodicities and clustering, annotate anomalies and produce a one-page figure set for the weekly meeting.
wiki
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Find and interpret rich data sources+
Search archival surveys, observatory logs and instrument telemetry for deep, multi-band records, extract and standardise candidate tables, then write a short memo interpreting which sources add value to our models.
esco
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Watch and assess3
Grow the practice1

What the work runs on

named inside the evidenced tasks
13 tasksApache Sparkperform scalable ETL, deduplication, and benchmarking across very large astronomy datasets
5 tasksApache Hiveperforms structured joins and creates versioned canonical tables from heterogeneous astronomical datasets
4 tasksAtlassian Confluencecollaboratively drafts and shares the report and embeds figures for review by the team
3 tasksApache Kafkastreams new observation batches into the training pipeline for online updates
3 tasksApache Airfloworchestrates ingest tasks, runs validation checks and alerts on schema or timestamp mismatches
2 tasksApache Hadoopstore and manage the distributed file system holding the 100 TB of raw and processed data
2 tasksAlteryxprototype sampling schemes and compute sample size and stratification metrics with repeatable workflows
1 taskAtlassian JIRAturn agreed actions into tracked tickets with owners and deadlines
Browse every evidenced software task →

The same task, four heights

this page is height one
ExecuteDo today's task, with fewer mistakesyou are here → ImproveMake it easy for the next person to acceptin the atlas → DecideWork out the right move when it is unclearin the atlas → BecomeLearn the pattern so it stops coming backin the atlas →

Can AI actually do this job?

the honest answer

It can

where it genuinely helps
  • Explain the theory behind the work
  • Draft, tidy and structure your writing
  • Rehearse a hard conversation before you have it
  • Build a study plan that fits your gaps

It cannot

where it stops, completely
  • Be in the room where a data scientist astronomy actually works
  • Carry the responsibility when the call is wrong — that weight stays yours
  • Notice what no one wrote down: the hesitation, the thing left unsaid
  • Live with the outcome

What the work pays

two countries, two different measures

United States

this exact occupation · BLS 2025
  • $120,230 a year — the middle: half earn more, half earn less
  • The lowest tenth earn near $67,240; the top tenth near $199,130
  • 262,440 people employed in this occupation

India

the occupation GROUP, not this job · PLFS via ILOSTAT 2025
  • ₹38,298 a month — the median for Professionals, the group this work sits in
  • India publishes pay by broad occupation group, so this covers many jobs besides this one. It is a shape, not a salary.
read this carefullyThese two numbers are not comparable and must not be converted into each other. One is a yearly figure for this job alone; the other is a monthly figure for a whole family of jobs. What travels between them is the pattern, not the amount: experience lifts pay almost everywhere.

Where the evidence lives

open any of it yourself
O*NETThe US Department of Labor's own record of this job9 of these tasks ESCOThe European skills and occupations register8 of these tasks Job descriptionsReal postings, in the wild14 of these tasks WikipediaHow the world describes this work5 of these tasks US Bureau of Labor StatisticsPay and employment, by occupation2025 figures

Close to this work

12 nearby
AstronomyPlanetary Scientist26 evidenced tasks AstronomySpacecraft Operations Engineer25 evidenced tasks AstronomyAdaptive Optics Engineer25 evidenced tasks AstronomyScience Journalist25 evidenced tasks AstronomySpace Mission Analyst23 evidenced tasks AstronomyMechanical Technician Observatory23 evidenced tasks AstronomyYoutube Space Educator23 evidenced tasks AstronomySpace Scientist22 evidenced tasks

Questions people actually ask

You’ll spend mornings running and checking pipelines that ingest telescope or simulation data using tools like Apache Airflow, Spark, or Hadoop. Expect to clean raw data, merge sources (catalogs, telemetry) and ensure dataset consistency before analysis.

Afternoons often go to developing and testing models (Python, Spark MLlib), visualizing results with dashboards, and meeting cross-functional teams — instrument scientists, software engineers, and project managers — to turn a science question into code and plots.

Common backend systems are Apache Spark and Hadoop for large-scale processing, Apache Hive for queryable data, and Kafka for streaming telescope telemetry. Airflow schedules workflows; Alteryx may be used for ETL (extract, transform, load).

Teams track work in Atlassian JIRA and write documentation in Confluence. Expect Python and SQL for coding, and plotting libraries or dashboards for visualizations.

You train models on labeled observations or simulations, then monitor performance with holdout tests and validation metrics to avoid overfitting. Use explicit uncertainty estimates — e.g., probabilistic models or calibration curves — because false positives can waste telescope time.

Document assumptions in Confluence, version datasets and models, and run reproducibility checks (same input -> same output). Share results with domain experts before actioning them, so instrument limits and selection biases are considered.

The U.S. Bureau of Labor Statistics (BLS) reports 262,440 employed data scientists and a median wage of $120,230 per year. The lowest tenth earn about $67,240, and the top tenth about $199,130, according to BLS 2025 data.

Salaries vary with employer type: universities and observatories may pay less than private space companies or large tech firms, and location and experience change where you fall in that range.

Learn Python and SQL first; practice with small datasets (FITS files, CSVs) and Jupyter notebooks. Take introductory statistics and a basic machine learning course. Play with Astropy and simple Spark locally or on cloud free tiers.

Work a small project: download public survey data (e.g., SDSS), clean and merge catalogs, build a classifier or clustering model, and make visualizations. Put code on GitHub and write a short Confluence-like README.

Statistical thinking combined with coding: you must know how sampling, bias, and uncertainty affect results and code that logic reliably. That means being able to design surveys or sampling strategies, clean and merge data, and interpret model outputs.

Practical tools matter too — proficiency with Spark or Hadoop for big data, plus version control and documentation (JIRA/Confluence) — but the core is being able to turn noisy telescope data into defensible scientific conclusions.