Marketing Decision Scientist — D2C & Agentic Commerce
iCustomer is the Agentic Decision OS for Growth Marketing, used by 100+ clients including American Eagle, Cisco, UNTUCKit, and Crusoe to turn audience intelligence into decisions and orchestration with learning loops built in. We're a venture-backed, US-headquartered startup with a growing India team, building the next category in growth marketing — the decision layer and helping growth teams optimize ROAS, CAC, and LTV. Customers don't just buy software from us they buy outcomes.
The Role We're hiring a Marketing Decision Scientist / Analyst with deep Shopify and D2C ecommerce chops to be the analytical brain behind our customer engagements. You'll work across our D2C book of business — UNTUCKit, American Eagle, and others — running the analyses that drive merchandising, media mix, and lifecycle decisions. You'll do it AI-natively: notebooks, agents, LLMs in the loop, and a point of view on where agentic commerce is taking the category. This isn't a dashboard role. You'll own the full arc from raw data → EDA → models → recommendations → measured lift. Your work shows up in customer QBRs, in our product, and in the decisions iWorkers make every day.
What You'll Do Analytics & Modeling Run market basket analysis, cohort analysis, RFM segmentation, and CLV modeling on Shopify and warehouse data. Build attribution models — MTA, MMM-lite, and incrementality tests that tie ad spend to revenue at the SKU and audience level. Build predictive models like propensity, churn, and next-best-action for D2C lifecycle programs. Design and analyze A/B tests and holdouts for paid media, email/SMS, and on-site personalization.
Notebooks & EDA Live in Jupyter / Colab / Hex / Claude Code. Build reproducible notebooks that engineers and PMs can read and ship from. Pull from Shopify, BigQuery, Snowflake, GA4, Klaviyo, and ad platforms — clean, model, visualize, narrate. Turn EDA into one-page customer-facing insights not 40-slide decks.
AI-Native Approach Use LLMs and agentic workflows as part of your daily toolkit — code generation, data exploration, hypothesis testing, narrative writing. Build evaluators and feedback loops for iCustomer's iWorkers (our role-based digital twins) on D2C use cases. Push the boundary on what an AI-native data scientist looks like in 2026. Agentic Commerce Develop a POV on where agentic commerce is heading — AI shopping agents, agent-to-agent purchase flows, conversational commerce, and what it means for D2C brands. Help shape iCustomer's roadmap for serving D2C brands in the agentic commerce era. Customer Impact Embed with our D2C customers' growth and analytics teams. Translate their messy data into clear decisions. Co-author case studies and Substack posts with the founder when the work is good enough to show.
Who You Are 2–6 years as a Data/Decision Scientist, Marketing Data Scientist, or Marketing/Analytics Analyst at a D2C brand, ecommerce platform, agency, or martech company. Shopify fluency — you know the data model (orders, line items, customers, variants, draft orders) and the quirks. You've worked with Shopify Plus, Shop Pay, or the Storefront API. Strong in Python (pandas, scikit-learn, statsmodels) and SQL (BigQuery or Snowflake preferred). Hands-on with attribution and incrementality — MTA, MMM, geo-experiments, holdout design. Comfortable with the modern D2C stack: GA4, Klaviyo, Meta Ads Manager, TikTok Ads, Northbeam / Triple Whale, Segment / RudderStack. AI-native you use Claude, ChatGPT, Cursor, or similar as part of how you work, not as a novelty. Strong written communicator. You can turn a model output into a recommendation a CMO can act on. Curious about agentic commerce, agent-to-agent transactions, and conversational shopping you've at least read and thought about it.
Bonus Points Curious about building SLMs (Small Language Models) — fine-tuning, distillation, RAG, evaluation. We'd love to give you a sandbox. Experience with dbt, Airflow / Prefect / Dagster, or modern data stack tooling. Built causal inference models (DoWhy, EconML, CausalImpact, synthetic control). Worked with CDPs or identity resolution (Segment, Bloomreach, Hightouch or open source tools) Published a notebook, blog post, or talk on D2C analytics or AI-native data work. B2B SaaS or B2B GTM analytics experience as a complement to D2C.
Logistics Location: Bangalore-based, remote with hybrid work expected (in-person collaboration days at the Bangalore hub). Hours: Meaningful overlap with US Eastern Time (~5 hours daily minimum, including customer-facing calls). Type: FTE Compensation: [Competitive INR comp + performance bonus + equipment + GPU/notebook credits] Reports to: Iqbal Kaur, Head of Decision Science
How to Apply Send us: Fill out this form: https://icustomerai.notion.site/03566926971183e7a82281e3bf1764ed?pvs=143