The posting
EverC, now part of G2 Risk Solutions, is a pioneer in its field. By leveraging technologies that illuminate the darkest corners of the internet and provide unparalleled visibility into the largest source of data in the world, EverC uses artificial intelligence and machine learning techniques to assess hundreds of millions of domains and effectively categorize the internet. These insights reveal hidden relationships and risks, identify an entity’s full digital fingerprint, and create new opportunities for businesses to scale efficiently and confidently.
About the Role
We’re looking for a hands-on Data Scientist who owns AI and LLM-powered products end to end — and cares about how they perform for real users.
You will own the modeling and evaluation of our fraud and risk detection products: exploring the data, designing models and prompts, evaluating them offline, and monitoring them in production. You’ll work side by side with engineers to ship your models into large-scale, real-time systems — you drive the data science and the product outcomes, while engineering owns the platform plumbing.
This is a data science role at heart. You won’t be a data engineer, but you’ll be comfortable working close to the data and the systems, and knowing your way around modern ML and data infrastructure is a strong plus.
Responsibilities
Data Science & Modeling
- Own the design, build, and evaluation of LLM-based and classical ML models for classification, entity extraction, and risk scoring on large-scale, real-world data.
- Lead prompt design and iteration — model configurations, fallback strategies, and tradeoffs between cost, latency, and quality.
- Apply sound statistical judgment on adversarial, noisy, or imperfectly labeled data.
- Drive the modeling decisions behind how LLMs and ML models fit into streaming, event-driven production systems.
Product & Platform Partnership
- Partner with engineers to ship your models into real-time streaming and large-scale batch pipelines — you own the modeling, they own the platform.
- Shape internal evaluation tooling, LLM observability, and model-performance monitoring.
- Define the requirements your infrastructure must meet to support rapid experimentation without hurting production reliability.
Data Quality & Product Performance
- Own the data-quality and product metrics that matter (precision, recall, coverage, latency, cost efficiency).
- Build measurement frameworks for production systems and offline experiments.
- Analyze production data to find labeling gaps, false-positive patterns, and new detection opportunities.
- Drive experimentation (A/B tests, shadow deployments, offline evaluation) and back every decision with measurable business outcomes.
Requirements
- 5+ years applied data science / ML, including production model deployment.
- Hands-on LLM prompt engineering and evaluation (not just fine-tuning theory).
- Strong Python.
- Experience with large-scale data processing concepts and technologies.
- Strong grasp of data-intensive applications and modern data architectures.
- Experience building cloud-native systems (AWS preferred).
- Excellent communication and stakeholder management.
- Comfortable with ambiguous, adversarial data and designing labels/heuristics where ground truth is imperfect.
Preferred Qualifications
A data scientist who is also fluent in the ML and data-engineering stack is a big plus:
- Data & Infrastructure: Apache Spark, Kafka, Kubernetes, Docker, EMR, Airflow, Iceberg / Delta Lake, K8s
- Cloud & DevOps: Infrastructure as Code (Terraform), CI/CD (GitHub Actions), monitoring & observability platforms
- AI & ML: ML platforms, LLM-powered applications, vector databases
