Role SummaryWe are hiring a AI Engineer to serve as a technical architect for our Intelligence Layer and Data Consortium. This is a specialized engineering role-distinct from general web development-focused on building the high-scale "muscle" that powers our fraud intelligence.
You will design, build, and operate distributed, production-grade services and data pipelines that ingest real-time signals from millions of users and enable our Agentic Flow to auto-tune strategies. You will own and evolve the internal AI agent workflow and tooling originally prototyped by our detection team, and help migrate it onto our new, production-grade agent framework. You will also play a key role in building AI applications and agentic flows using state-of-the-art, out-of-the-box large language models (LLMs), while partnering with Data Science and Solutions to integrate traditional machine learning models, rule engines, and label pipelines into production.
This role is first and foremost a production software engineering role. Classic ML modeling experience is a plus, but not required, as long as you bring strong software engineering fundamentals and a solid understanding of ML concepts.
Primary Responsibilities- Consortium Data Engineering
Architect and maintain high-throughput data pipelines (using technologies such as Spark, Kafka, or Flink) to ingest, process, and aggregate real-time signals-such as device fingerprints and behavioral biometrics-into our central intelligence graph.
- High-Scale System Design
Design and optimize distributed systems to support our global data network, ensuring the platform can handle 10,000+ Transactions Per Second (TPS) with P99 latency under 150ms.
- Agentic Flow & AI Application Development
Build agentic flows and AI applications by leveraging state-of-the-art, out-of-the-box LLMs (e.g., OpenAI, Anthropic, Google) to enable natural language interaction, intelligent rule merging, and automated fraud strategy recommendations.
- AI Agent Workflow Ownership
Own and extend the internal AI agent tool and workflows used by the Solutions team for rule and feature creation, rule tuning, and alert analysis, ensuring reliable deployments across sandbox, preprod, and production solution tenants.
- Label & Rule Tuning Automation
Build map-reduce style LLM workflows and analytics pipelines (e.g., ClickHouse, Spark) for large-scale label investigation, weak classifier discovery, and FN/FP triage to accelerate solution onboarding and improve detection coverage.
- Productionize ML Pipelines
Collaborate with Data Scientists to deploy and maintain pipelines for both Unsupervised (UML) and Supervised (SML) models, integrating them with our APIs to enable real-time scoring and decisioning. Hands-on ownership of classic ML modeling is a plus, but not a strict requirement.
- Privacy-First Architecture
Implement robust security measures, including tokenization and hashing, to ensure PII privacy and compliance across our shared intelligence network.
- Cross-Functional Collaboration
Work closely with Data Science, Product, Strategy, Delivery, and Engineering teams to develop, validate, and optimize machine learning-driven features and AI-powered workflows.
Requirements- Experience
2+ years of professional software engineering experience building and shipping production systems (backend services, data platforms, or ML/AI infrastructure), ideally for customer-facing SaaS products or internal platform tools.
- Education
Bachelor's and Master's degree in Computer Science (or a closely related field) with a focus in Machine Learning or Artificial Intelligence.
- System Architecture
Proven ability to design and implement distributed, cloud-native systems for high-throughput, low-latency applications. Experience with AWS and containerization (Docker/Kubernetes) is required.
- Coding Proficiency
Strong, production-grade skills in Python (primary language for services and tooling), plus experience with at least one additional lower-level programming language such as Java (Go or C++ also a plus).
- Big Data Technologies
Hands-on experience with distributed data frameworks such as Spark, Kafka, or Flink.
- Machine Learning Foundations
Solid breadth and depth in ML concepts (e.g., supervised vs. unsupervised learning, feature engineering, embeddings, evaluation metrics like Precision/Recall and AUC), even if you have not been the primary model owner on a team.
- Collaboration & Ownership
Demonstrated ability to work cross-functionally, take end-to-end ownership of services, and operate in a fast-paced, high-impact environment.
- Preferred Qualifications
- Experience building or integrating LLM-powered agent workflows (e.g., LangChain/LangGraph, multi-agent orchestration, tool calling, or RAG architectures) for production or internal platforms.
- Experience deploying or maintaining machine learning models (supervised or unsupervised) in production environments.
- Experience working with analytical databases and large-scale data exploration (e.g., ClickHouse or other columnar data stores) is a plus.
- Background in real-time decision engines or stateful stream processing.
- Domain knowledge in fraud or risk (fraud detection, credit risk, payments, or trust & safety) is a strong plus but not required.
Benefits- Base Salary: 130K - 250K
- Total Compensation: Includes Base + Performance Bonus + Equity Options.
- Comprehensive medical, dental, and vision coverage.
- Discretionary Time Off Policy (DTO) and paid holidays.
- Opportunities for R&D exploration and professional development.
- Regular team-building events and a collaborative, innovative culture.
