Claude Prompts for Data Scientist Job Search

You are an expert recruiter for data science roles. Here is my resume:

[PASTE RESUME]

And the job description:

[PASTE JOB DESCRIPTION]

Rewrite my bullet points to match this role. Lead each with the modeling outcome (a metric moved, lift over baseline, money saved, a decision shipped), quantify it, and mirror the exact tools and methods from the job description (for example Python, PyTorch, scikit-learn, SQL, experimentation, causal inference, feature engineering) without inventing anything. Flag any bullet that reads as a stretch.

Act as an ATS parser for data scientist roles. From this job description:

[PASTE JOB DESCRIPTION]

List the languages, libraries, modeling techniques, and exact keyword phrases (for example XGBoost, deep learning, A/B testing, Bayesian methods, MLflow, model deployment) an ATS would score against, ranked by prominence. Then tell me which are missing from my resume:

[PASTE RESUME]

and where I could add each one truthfully.

Help me quantify this data science project for my resume: [DESCRIBE PROJECT AND MODEL]. Ask me up to 4 questions to surface the impact (baseline vs model performance, what business metric it moved, whether it shipped to production, how many users or dollars it touched). Then write 2 resume bullets in the format 'accomplished X by doing Y, resulting in Z', keeping the modeling detail specific but not inflated.

Write a concise cover letter for this data scientist role:

[PASTE JOB DESCRIPTION]

Using my background:

[PASTE RESUME OR SUMMARY]

Under 250 words, no cliches. Open with a concrete reason I fit this team or problem space, cite one model or experiment I ran and the decision it drove, and close with a clear call to action. Match the company's tone.

Write a 4 to 5 sentence LinkedIn message to the hiring manager for this data scientist role: [PASTE ROLE AND COMPANY]. Introduce me briefly, give one specific reason I am interested in their team or the problem they work on, and mention one relevant model or analysis I have shipped ([DETAIL]). Warm, human, under 90 words.

Act as a senior data scientist interviewing me for a [SENIORITY] role. Ask me one realistic modeling question (for example: how would you build a churn model, or a recommendation system, or a fraud detector). After I answer, probe like a real interviewer: ask about feature selection, class imbalance, the metric I would optimize, overfitting, and how I would deploy and monitor it. Then tell me what a strong answer covers. Ask the question now and wait for my answer.

Quiz me on the statistics and experimentation questions common in data scientist interviews: p-values and their pitfalls, statistical power, sample size, multiple testing, confidence intervals, and how to design and read an A/B test (including novelty effects and peeking). Ask one question at a time, wait for my answer, then correct my reasoning and give the crisp version I should have said.

Act as a hiring manager running a data science case study. Give me one open business scenario (for example: revenue dropped 12% this quarter, or we want to reduce delivery times using data). Push me to clarify the goal, define the metric, list the data I would pull, choose a modeling or experimental approach, and state how I would measure success. After I answer, tell me where a strong candidate would go deeper. Ask the question now and wait.

Compare my resume against job descriptions for [TARGET ROLE, for example Senior ML-focused Data Scientist]. My resume: [PASTE RESUME]. Target job descriptions: [PASTE 2 to 3]. Identify the specific modeling, tooling, and domain gaps between me and these roles, ranked by how much they matter, and for each suggest the fastest credible way to close or reframe it.