Wiki

Job Descriptions

Archive-backed guidance for reading and writing data job descriptions: role clarity, problem framing, requirements, red flags, and candidate fit.

Definition and Scope

Job descriptions are the shared role spec between a hiring team and a candidate. They describe the problem and the role. They also name the level, required evidence, and hiring process. In data work, they also help disambiguate unstable titles such as Data Scientist, Data Engineer, and Data Analyst.

The DataTalks.Club hiring archive treats job descriptions as more than recruiting copy. They’re a diagnostic for team maturity. In Hiring Data Scientists and Analysts, Alicja Notowska describes the job spec as something recruiters build with hiring managers around 7:09. She then explains how market reality changes requirements around 13:57 and 17:18.

In Data Science Jobs, Tereza Iofciu gives the candidate side. When the title and responsibilities disagree, the description reveals that the company may not know which data problem it’s hiring for.

Common Definition

A useful data job description names the team and business problem. It states the level, responsibilities, must-have skills, and interview process. It should also separate essentials from preferences. Alicja’s recruiting episode makes that separation operational. Recruiters and hiring managers decide what the role needs, then check whether the market can supply those requirements.

The candidate-facing definition is similar. In Land Data Scientist Roles, Luke Whipps advises candidates to research the company problem around 37:54. He then maps the CV, portfolio, and outreach to that problem. In Data Science Interview Guide, Oleg Novikov tells candidates to use the job description around 15:29 and 17:13. The posting helps them infer whether the company wants product data science, machine learning engineering, analytics, or another role structure.

A job description is therefore both a hiring artifact and a reading artifact. Hiring teams use it to define the work. Candidates use it to decide whether the work matches their skills and whether the company understands the role.

Guest Differences

Alicja focuses on the recruiter and hiring-manager workflow. She emphasizes role alignment and sourcing in Hiring Data Scientists and Analysts. She also covers inclusive language, screening, and salary conversations. That makes the job description a tool for running the Hiring process.

Tereza is more skeptical in Data Science Jobs, where she frames vague descriptions as warning signs for candidates. Around 20:06 and 23:01, she reads titles and team context as evidence. Around 30:20, responsibilities and tool lists also show whether the company has a coherent Data Team.

Luke and Oleg mostly discuss job descriptions from the applicant side. Luke uses them for targeted applications, especially around 37:54 and 44:26 in Land Data Scientist Roles. Oleg uses them for interview prediction and CV positioning in Data Science Interview Guide. Nicolas Rassam adds a Data Engineering version in Hiring Data Engineers in Europe. Around 18:47, 22:55, and 31:16, he treats titles as noisy.

For data engineering roles, Nicolas looks for SQL and Python. He also looks for problems, outcomes, projects, and level-appropriate responsibility.

Role Requirements

Role requirements should describe the work before the tool stack. Alicja says job descriptions should focus on problems over perks around 18:28 in Hiring Data Scientists and Analysts. That means a data scientist description should clarify whether the person will run experiments, build models, support product decisions, or deploy ML systems.

A data analyst description should clarify whether the work is BI reporting, product analytics, stakeholder analysis, or analytics engineering. A data engineering description should name the operating surface. The work may involve pipelines, platform infrastructure, data models, or governance controls.

Requirements should also match level. Nicolas discusses junior-to-senior data engineering expectations around 22:55 in Hiring Data Engineers in Europe. Junior descriptions should leave room for training and mentorship. Senior descriptions can expect system ownership, architecture judgment, and cross-team communication.

When one posting asks for junior compensation and senior scope, the mismatch belongs on both Job Search and CV Screening. It will distort both applications and evaluation.

Hiring Signals

A strong description helps candidates show the right evidence. Alicja describes screening experience and education around 21:32 and 27:10 in Hiring Data Scientists and Analysts. She then discusses responsibilities and CV clarity around 28:41 and 32:40. Those signals work only when the description explains what evidence matters.

If the role needs experimentation, the description should name experiment design, metrics, and product decision work. If the role needs data engineering, it should name the pipeline work and the expected ownership. SQL, Python, cloud fundamentals, and data quality can then appear as evidence instead of a loose keyword list.

Luke’s candidate advice points in the same direction. Around 16:15, 19:50, and 25:04 in Land Data Scientist Roles, he looks for industry fit and use-case alignment. He also looks for concrete projects and business impact.

A job description that names the work helps candidates choose the right evidence. They may show a dashboard, a modeling case study, or a pipeline. For production roles, they may show a service or another portfolio artifact.

Oleg adds that candidates should treat the CV like a landing page around 18:28 in Data Science Interview Guide. That advice depends on the job description. A good posting gives the candidate enough signal to put relevant achievements first and remove unrelated details.

Job-Posting Pitfalls

A misleading title can hide a different role. Tereza’s Data Science Jobs episode calls this out around 20:06 and 23:01. A “data scientist” posting full of ETL, Airflow, and data platform work may be Data Engineering. A “data analyst” posting that owns instrumentation, dbt models, and semantic layers may be closer to Analytics Engineering.

Long technology lists are another warning sign. Tereza discusses overloaded tech lists and vague responsibilities around 30:20 and 33:33. Tools matter, but the description should explain why they matter.

Airflow often means batch pipelines, while dbt often means analytics engineering. A vector database may mean search or RAG. Without that context, the tool list becomes keyword noise.

Alicja discusses inclusive job-description wording around 20:04 in Hiring Data Scientists and Analysts. Tereza flags words such as “rockstar” and “ninja” around 31:03 in Data Science Jobs. The issue isn’t style alone because such wording can signal unclear expectations, hero culture, or a narrow view of who belongs in the role.

Missing salary context and unclear interview steps also weaken a description. Tereza discusses salary transparency around 37:08. Alicja covers salary bands and negotiation around 40:33 in Hiring Data Scientists and Analysts. When a posting hides level and compensation, candidates have to discover basics during recruiter calls. The same problem appears when the posting hides process or team context.

Portfolio Evidence

Portfolio evidence should answer the job description, not display unrelated work. Luke tells candidates to connect projects to concrete use cases around 19:50 and business impact around 25:04 in Land Data Scientist Roles. Oleg recommends cold-start projects and synthetic data for candidates without direct industry experience around 45:46 in Data Science Interview Guide. He also suggests blogging as a way to make that work visible.

Nicolas gives the data-engineering version around 54:25 and 55:53 in Hiring Data Engineers in Europe. Shareable projects and GitHub work can show pipeline thinking, privacy awareness, and clear storytelling.

Portfolio work should prove the role requirement. For Data Engineering Portfolio Projects, use ingestion and orchestration as evidence. Add tests, data quality, and a runbook.

For Machine Learning Portfolio Projects, use model framing, evaluation, and deployment. Add monitoring and tradeoff explanation.

For Analytics Engineering Portfolio Projects, use clean models and metrics definitions. Add stakeholder-facing documentation and decision support.

Related Pages

Use Hiring for the employer-side process, Job Search for candidate strategy, and CV Screening for how recruiters read applications. Role-specific context lives in Data Scientist Role, Data Engineer Role, Data Analyst Role, and Data Teams.