AI Engineer

About Us

Xactus (pronounced ‘Zac-tus’) is the leading verification innovator for the mortgage industry. We have over 6,500 clients ranging from the largest bank and non-bank mortgage originators to credit unions and mortgage brokers. Xactus works closely with our clients to digitally integrate a 360° approach to verification across their workflows. As a result, lenders can easily access the technology necessary to meet consumer demands for a modern mortgage experience with industry-leading speed, reliability, and accuracy – while also closing more loans more quickly with greater profitability.

Xactus is proud to provide a friendly work environment that is primarily remote. Our workforce offers many opportunities to enhance your skills with our top-notch financial leadership team who prioritizes building talent.

About You

You are a career-minded, driven individual who is looking for a position that challenges you and supports your professional development.

Responsibilities

• GenAI Feature Development: Build production GenAI applications using LangChain, LangSmith, and Agentcore. Implement complex agent workflows with tool use, memory management, and multi-step reasoning. Design and optimize agentic patterns for reliability, latency, and cost.
• RAG & GraphRAG Architecture: Design and implement retrieval-augmented generation systems using semantic search and vector databases. Architect knowledge graphs using Neo4j/Neptune for structured reasoning. Optimize retrieval strategies for accuracy, latency, and relevance ranking.
• Data Pipeline & ETL Development: Write custom Python ETL scripts for data preparation, indexing, and synchronization. Design schema and data models for graph databases. Collaborate with data team on AWS Glue orchestration and S3 management.
• Database & Query Optimization: Write optimized SQL queries and design database schema. Optimize Neo4j/Neptune Cypher/SPARQL queries and implement efficient vector similarity search in OpenSearch.
• Backend API Development: Build production FastAPI services for GenAI features. Implement async request handling, error management, and fallback strategies for LLM unreliability.
• Model Tracking & Experimentation: Establish experiment tracking using MLflow, including model versioning, hyperparameter logging, and metrics comparison. Log and monitor LLM outputs for quality and bias.
• Production Monitoring & Debugging: Implement logging, metrics, and tracing using CloudWatch and LangSmith. Monitor LLM latency, token usage, and costs. Respond to production incidents with rapid remediation.
• Code Quality & Testing: Write unit, integration, and end-to-end tests. Perform code reviews and test AI system outputs for correctness and safety.
• Cross-Functional Collaboration: Partner with DevOps to deploy GenAI services. Communicate with data engineering on pipeline needs. Work with product to translate requirements into technical designs.
• Documentation & Knowledge Sharing: Document system architectures, RAG/GraphRAG designs, and operational runbooks. Share learnings on agent patterns and prompt engineering.

Requirements

• Bachelor’s degree in related field or equivalent.
• 3–5 years building production AI/ML systems or full-stack applications.
• Experience with ML/AI frameworks (PyTorch, TensorFlow, scikit-learn).
• 1–2+ years with LangChain, LangSmith, or similar LLM frameworks.
• Hands-on experience with Claude and/or GPT models.
• 2+ years of AWS experience (S3, CloudFormation, CloudWatch).
• Strong Python and FastAPI.
• SQL expertise.
• RAG and semantic search experience.
• Software engineering discipline (testing, code review, CI/CD).

Preferred Qualifications

• AWS Bedrock.
• Codex/code generation models.
• Prompt engineering at scale.
• GraphRAG/Neo4j/Neptune.
• Deep learning (CNNs, RNNs, transformers).
• NLP (tokenization, embeddings, text classification).
• Hyperparameter optimization.
• Computer vision/OCR.
• Terraform, Docker/Kubernetes.
• MLflow, OpenSearch, AWS Glue.
• Track record of shipping GenAI features.

Skills

• Strong written and verbal communication skills.
• LLM Mastery: Deep understanding of Claude, GPT, and Codex capabilities, limitations, and failure modes.
• Problem Solving: Breaks complex AI problems into components; designs experiments to validate strategies.
• Production Reliability: Builds for failure, monitors proactively, responds rapidly to production issues.
• Cross-Functional Collaboration: Communicates effectively with DevOps, data engineers, and product teams.

Benefits

• Medical, vision, and dental insurance.
• Bonus programs.
• Fitness reimbursement and other healthy lifestyle programs.
• 401k plan with a company match.
• Short and long-term disability, life insurance, accident, and critical illness insurance.
• Health savings account and flexible spending account.
• Employee assistance program, legal services, and employee discounts.

A friendly, supportive environment which is highly rated by Xactus employees. Feedback from our employees includes: “The people I work with treat each other with respect,” “I feel accepted by my coworkers,” and “The person I report to cares about me as a person.”

Working Conditions

Traditional office environment with low-to-moderate office noise (computers, phones, and business conversations). The position may be remote from main offices. May require flexibility in hours.

Physical Demands

• Lifting/carrying up to 10 lbs.
• Manual dexterity for computer work.
• Speaking, hearing, and vision are required to perform essential functions.

Xactus promotes an equal opportunity workplace, which includes reasonable accommodations of otherwise qualified disabled applicants and team members.