Your First 90 Days as a Data Scientist

2026-02-14 · Source: Towards Data Science · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, medium

Summary

A Data Science Manager at DoorDash shares a structured approach for the first 90 days in a new data science role, emphasizing building connections, domain understanding, and data knowledge. The onboarding process is broken down into three phases: Foundations (Weeks 1-2), Getting Hands-On (Weeks 2-6), and Ownership (Weeks 6-12). Key strategies include frequent meetings with managers and cross-functional partners, leveraging AI tools like Glean and NotebookLM for domain context, and utilizing AI-assisted data tools such as Cursor for SQL query generation. The author also highlights the importance of setting up the tech stack early, understanding key metrics, and making early contributions like improving documentation or suggesting process enhancements to establish trust and ownership.

Key takeaway

For Data Scientists or Managers starting a new role, prioritize establishing cross-functional alignment, business fluency, and data intuition within your first 90 days. Focus on building connections, leveraging AI tools for rapid domain and data knowledge acquisition, and making small, early contributions to build trust and ownership. This structured approach will accelerate your ramp-up and position you for greater impact.

Key insights

Successful data science onboarding prioritizes connections, domain context, and data knowledge to drive business impact.

Principles

• Build connections to embed deeply in the business.
• Understand business context to influence decisions.
• Hands-on data work provides practical guidance.

Method

Onboarding involves three phases: establishing foundations (weeks 1-2), hands-on project work and deep dives (weeks 2-6), and achieving ownership and influence (weeks 6-12).

In practice

• Use AI tools (e.g., Glean, NotebookLM) for document summarization.
• Leverage AI-assisted data tools (e.g., Cursor) for SQL query generation.
• Contribute early by improving documentation or suggesting process changes.

Topics

• Data Science Onboarding
• AI-Assisted Workflows
• Domain Knowledge Acquisition
• Data Knowledge
• Cross-functional Collaboration

Best for: Data Scientist, Machine Learning Engineer

Related on AIssential

• Where to Start Learning Data Science in 2025 — Foundational data science skills are attainable in 30 days by focusing on practical tools and a structured learning path.
• Breaking Into Data Roles in the Age of AI: What Actually Matters Today — Modern data roles demand practical system understanding, cloud proficiency, and collaborative skills beyond theoretical tool knowledge.
• The Best Data Scientists Are Always Learning — Sustained learning requires personalized incentives, structured progress, reduced pressure, and dedicated reflection time.
• What Happens When a Beginner Refuses to Stay a Beginner — Practical project experience is crucial for understanding real-world machine learning challenges beyond structured course material.
• You're learning data science. You're not becoming one. — The gap between learning data science and becoming a practitioner is bridged by real-world "friction" that builds judgment.
• Build a Resume That Lands AI Engineering and Data Science Interviews - Only 10 Seats Left — The workshop provides practical guidance and resources for optimizing resumes for AI/data science roles.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Towards Data Science.