Shape the future of client intelligence by building a robust, data-driven system from the ground up. In this role, you'll architect and maintain a comprehensive market intelligence database that powers sales decision-making. Your work will directly influence how we identify and prioritize high-potential nonprofit clients, using advanced modeling and real-time data signals.
What You’ll Do
- Develop and maintain scalable web scrapers to gather behavioral and technical signals from nonprofit websites, including payment tools, hosted products, and sector-specific markers.
- Design and train machine learning models—focusing on classification and uplift modeling—to score prospects and improve lead qualification accuracy.
- Integrate diverse data sources, including nonprofit registries, web traffic analytics, and social signals, into a unified, enriched dataset.
- Apply strong statistical reasoning to connect model performance (like AUC or F1) with business outcomes such as conversion and customer lifetime value.
- Structure and store processed data in ClickHouse and MongoDB, ensuring accessibility and reliability for downstream teams.
- Collaborate with sales to analyze disqualifications, uncover patterns, and iteratively refine model logic to reflect real-world feedback.
- Deploy scoring pipelines into production using Docker and CI/CD workflows, with outputs seamlessly flowing into Salesforce.
- Monitor client sites for correct implementation of core tools through automated change detection and tracking systems.
What We’re Looking For
- At least 5 years of experience applying machine learning to solve tangible product challenges.
- Deep familiarity with gradient boosting, statistical modeling, and evaluation metrics across classification and regression tasks.
- Proven ability to design and manage large-scale data pipelines, including scraping infrastructure and data normalization.
- Strong engineering fundamentals: clean, maintainable Python code, solid grasp of design patterns, and experience with MLOps tools like MLflow and Airflow.
- Proficiency in SQL and experience building complex datasets in ClickHouse and MongoDB.
- Independence in problem-solving—able to define scope, select appropriate technologies, and drive projects to completion.
Preferred Traits
- A hypothesis-driven approach to data exploration and model development.
- Ability to translate technical findings into clear insights for non-technical stakeholders.
- High attention to detail and a strong sense of ownership over end-to-end systems.
- Comfort navigating ambiguity in fast-moving, data-intensive environments.
Technology Environment
You’ll work with Python (using uv and ruff), FastAPI, Pydantic, Docker, CatBoost, CausalML for uplift modeling, OpenAI via RAG and prompt engineering, ClickHouse, MongoDB, pandas, Polars, Redis, MLflow, Airflow, Grafana, Sentry, and Linux server administration. Distributed computation and CI/CD are core to our workflow.
Work & Culture
This is a remote role with team members across locations. We operate with a flat structure—no bureaucracy, just meaningful work and open communication. You’ll join a product-focused team that values transparency, long-term thinking, and shared learning. Equity is offered to reflect sustained contribution, and we support continuous growth through education, workspace setup, health benefits, and flexible time off.
