Role Overview
The Data Scientist, Next Gen Recommendation Systems role at impact.com in New York City involves building and scaling advanced recommendation frameworks for a global commerce partnership platform. This position suits a skilled data scientist with a background in applied machine learning who wants to design end-to-end production systems that connect advertisers, publishers, creators, and consumers.
Key Responsibilities
Design, build, and evaluate multi-entity recommendation models spanning candidate generation, ranking, reranking, and personalization across the partnership graph.
Evolve the architecture toward graph-based approaches by learning semantic embeddings of entities and relationships, and applying graph neural networks or attention-aware graph transformers.
Build models and pipelines for both batch and real-time serving, partnering with engineering on vector search, feature stores, and low-latency infrastructure.
Own the full model lifecycle from data and feature design through to production deployment, monitoring, and iterative improvement.
Leverage AI coding agents daily to accelerate prototyping, refactoring, debugging, and shipping production code.
Design offline evaluations and online experiments like A/B tests and holdouts while managing common pitfalls such as position bias and feedback loops.
Collaborate closely with product, engineering, and business stakeholders to translate platform goals into measurable model outcomes.
Requirements
3+ years of professional experience in data science or applied machine learning with a track record of shipping production models.
Strong proficiency in Python and SQL, alongside experience working with large-scale data and distributed compute frameworks like Spark or Databricks.
Hands-on experience developing recommendation or ranking systems, including candidate generation, learning-to-rank, or implicit feedback modeling.
Familiarity with embeddings and representation learning for users, items, content, or other ecosystem entities.
Demonstrated capability in machine learning engineering to build, ship, and maintain production pipelines rather than just notebook prototypes.
Strong experimentation skills encompassing the design, execution, and analysis of A/B tests.
Relentless daily use of AI coding agents as a workflow accelerator and an insatiable curiosity for learning new techniques quickly.
Nice to Have
Experience with graph-based machine learning such as graph neural networks, graph-aware retrieval, or knowledge graph embeddings.
Familiarity with modern deep learning recommender architectures including two-tower models, sequence-based recommenders, and multi-task ranking.
Practical knowledge of vector search and vector databases like FAISS, ScaNN, Vespa, or Milvus.
Background in real-time machine learning serving, feature stores such as Feast or Tecton, and low-latency inference patterns.
Exposure to contextual bandits, reinforcement learning, or off-policy evaluation settings.
Familiarity with PyTorch, TensorFlow, PyTorch Geometric, or DGL.
Experience working with GCP tools like Vertex AI, BigQuery, and Cloud Run, or mature MLOps practices.
Background in adtech, martech, e-commerce, or multi-sided marketplace recommendation systems.
Skills and Tools
Python
PyTorch
TensorFlow
SQL
GCP
Spark
MLOps
Salary and Benefits
Annual salary range of $100,000 to $125,000.
Additional 5% variable annual bonus contingent on company performance.
Restricted Stock Unit (RSU) grant with a 3-year vesting schedule.
Comprehensive medical, dental, and vision insurance coverage.
Responsible paid time off (PTO) policy for flexible rest and recharge.
Mental health and wellness benefits providing up to 12 fully covered therapy or coaching sessions per year alongside a monthly gym reimbursement.
Parental support including 26 weeks of fully paid leave for primary caregivers and 13 weeks for secondary caregivers.
Technology stipend for home office setup and a monthly internet allowance.
Office-only catered lunches every Thursday, healthy snack bars, and coffee.
About the Company
impact.com is the world's leading commerce partnership marketing platform that transforms business growth by enabling brands to discover, manage, and scale partnerships across the customer journey. Over 5,000 global brands, including Walmart, Uber, Shopify, Lenovo, L'Oréal, and Fanatics, rely on the company to power performance-based growth through integrated affiliate, creator, and advocate products.
How to Apply
Please apply directly on the official company careers page to submit your application for this position.
