Sev1Tech is seeking an AI Integration Engineer to integrate AI models into production systems, ensuring robust performance, real-time monitoring, and secure operations.
Requirements
- Proficiency in Python; knowledge of JavaScript, C++, or Go is a plus for UI or system-level integration.
- Experience with containerization (Docker, Kubernetes) and API development (REST, GraphQL).
- Expertise in logging frameworks (e.g., ELK Stack, OpenTelemetry) and visualization tools (e.g., Plotly, Chart.js).
- Understanding of AI model metrics (e.g., F1 score, latency) and drift detection techniques (e.g., PSI, KS test).
- Knowledge of AI vulnerabilities (e.g., prompt injection, model inversion) and mitigation strategies (e.g., differential privacy, ART).
- Experience with dashboarding tools (e.g., Grafana, Kibana) and observability platforms (e.g., Prometheus, Datadog).
- Familiarity with cloud platforms (e.g., AWS, Azure, Google Cloud) for AI deployment.
Responsibilities
- Model Integration: Integrate AI/ML models into applications (e.g., web, mobile, IoT) using APIs (REST, gRPC) and platforms like TensorFlow Serving or AWS SageMaker.
- Dashboard Development: Create real-time and historical dashboards using Grafana, Kibana, or Plotly to monitor model health (e.g., latency, accuracy) and data drift.
- Drift and Health Monitoring: Implement monitoring pipelines with tools like Evidently AI or Weights & Biases to detect data drift and model degradation, triggering alerts as needed.
- Logging and Tracing: Set up logging systems with ELK Stack, OpenTelemetry, or LangSmith to capture AI events, errors, and traces for debugging and auditing.
- Security Implementation: Apply secure-by-design principles to protect models and data from vulnerabilities (e.g., adversarial attacks, data leakage) using tools like Adversarial Robustness Toolbox (ART).
- System Optimization: Optimize model inference for performance (e.g., via quantization, edge deployment) and ensure compatibility with cloud (AWS, Azure) or on-premises infrastructure.
- Testing and Validation: Perform end-to-end testing of AI integrations, including stress testing and validation of dashboard metrics.
Other
- Bachelor’s or Master’s degree in Computer Science, Software Engineering, Data Science, or a related field.
- 4+ years in software engineering or AI integration, with experience deploying AI models in production.
- MUST BE US CITIZEN FOR GOVERNMENT CLEARANCE OR CONTRACT REQUIREMENTS
- Strong problem-solving skills for debugging integration issues and optimizing dashboards.
- Excellent communication to translate technical metrics into business insights.
