Senior Data Scientist

Job Description

Senior Data Scientist

Experience: 4+ years of applied data science delivery

USA - Remote

About the Role

You will work as a forward deployed data scientist: embedded with enterprise clients, turning ambiguous business problems into working analytical and AI solutions, and staying with the work until it runs in their environment. This is a hands-on senior role. You will own problems end to end — scoping with stakeholders, wrangling messy client data, choosing methods that fit the constraints, building the solution, and handing it off in a state another team can operate. Some engagements need something built from scratch. Many do not — the work is adopting, configuring and extending what already exists: our own products and accelerators, the client’s current platform, or a vendor tool. We are looking for people who are comfortable either way, who can tell which situation they are in, and who would rather extend a working solution than rebuild it. We are looking for modern technical builders rather than notebook-only analysts: fluent with Python, SQL, Git, cloud and cloud data platforms, AI-native in how they work, credible in front of a client, and pragmatic about when a simple method is enough and when deeper modelling is warranted.

Responsibilities

• Partner directly with client stakeholders to translate business problems into analytical problems — and push back when the framing is wrong.
• Build reliable analytical assets on messy enterprise data: forecasting, optimization, ML, statistical models, and the pipelines around them.
• Just as often, work with what is already there — configure and extend our existing products and accelerators, or the client’s existing tools, rather than rebuilding from scratch.
• Put something in a stakeholder’s hands early — a prototype app, a dashboard, a clear analysis — and iterate from real feedback.
• Ship code others can reuse tested, documented, reproducible, reviewed.
• Work alongside data engineers and ML engineers through to deployment, rather than throwing work over the wall.
• Use LLMs and AI coding tools to move faster, while keeping statistical judgment and final review your own.
• Raise the level of the people around you: review work and mentor less experienced colleagues.

Qualifications

Data Science

• Strong practical data science skills, including working with messy enterprise data and translating business problems into actionable analysis.
• Working knowledge of forecasting, optimization, machine learning, and statistical methods — with the judgment to know when a simple method is sufficient, and the discipline to validate honestly (proper back testing and out-of-time validation for time series, not just a random train/test split).
• Able to answer, “did this actually work?” with defensible evidence: experiment design and read-out, or quasi-experimental methods when a clean test is not possible.
• Regular user of LLMs or AI coding tools for coding, analysis, debugging, testing, and documentation, while maintaining independent judgment. You should be able to explain what AI output you accepted, rejected, rewrote, and tested, and how you spot output that is plausible but wrong.
• Basic understanding of production data/ML concepts, including pipelines, data quality, reproducibility, versioning, deployment, and monitoring.

Required Skills

• Strong proficiency in Python and SQL, at a level where your code goes to other people rather than only into your own notebook.
• Comfortable with Git, pull requests, code review, terminal workflows, package management, and reproducible environments.
• Working experience on at least one major cloud (AWS, Azure, or GCP) and with a modern data platform such as Snowflake, Databricks, Big Query, Redshift, Spark, or equivalent.
• Able to turn exploratory analysis into reusable, tested, and documented code.
• Comfortable working inside an existing codebase or product as well as starting from a blank page — reading someone else’s code, understanding how it works, and extending it without breaking it.
• Able to independently troubleshoot common data, query, environment, and pipeline issues — failed jobs, broken queries, bad joins, package conflicts, data-quality problems — without immediately requiring an engineer.

Preferred Skills

Delivery and client work

• Able to take an ambiguous request from a business stakeholder and turn it into a scoped piece of work, with a view on what is feasible in the time available.
• Able to clearly communicate assumptions, limitations, findings, and recommendations to both technical and business stakeholders — in writing and in the room.
• Able to make results usable: a lightweight app or dashboard (for example Streamlit, Dash, Plotly, or a BI tool) or a clean, well-argued document, depending on what the audience needs.
• Comfortable working in someone else’s environment and under their constraints — client tooling, access restrictions, and shifting priorities — without stalling.
• Strong engineering practices, experience with advanced LLM workflows, and experience delivering in enterprise environments are an asset.