AI Engineer

The AI Engineer is responsible for designing, developing, and deploying production-grade Artificial Intelligence solutions. The role focuses on leveraging Generative AI (LLMs) and Machine Learning to solve complex logistics and freight management challenges. Beyond model development, this position requires a strong grasp of MLOps to ensure that AI models are scalable, monitored, and seamlessly integrated into the enterprise ecosystem via APIs and middleware.
MLOps & Infrastructure
• Deployment & Orchestration: Containerize AI services using Docker and manage large-scale deployments on Kubernetes clusters.
• Pipeline Automation: Build and maintain robust CI/CD pipelines for automated model testing, deployment, and versioning.
• System Integration: Collaborate with System Architects to integrate AI modules into internal software and third-party vendor systems (Accounting, GPS, Freight Management) via APIs and Enterprise Service Bus (ESB) architectures.
Data Strategy & Monitoring
• Vector Infrastructure: Design and manage Vector Databases (e.g., Pinecone, Milvus, or Weaviate) to support semantic search and GenAI capabilities.
• Performance Monitoring: Develop telemetry dashboards to monitor model health, data drift, and hardware utilization (CPU/GPU/RAM).
Model Development & Innovation
• Architect AI Solutions: Design and implement Generative AI applications, specifically utilizing RAG (Retrieval-Augmented Generation) and Agentic workflows.
• Full-cycle ML: Lead the end-to-end development of ML models, from data collection and preprocessing to training, validation, and hyperparameter tuning.
• Model Optimization: Fine-tune open-source models (e.g., Llama, Mistral) and optimize inference speed for low-latency, real-time logistics applications.
• Model Reliability & Trust: Ensure AI outputs are accurate, grounded, and minimize "hallucinations" to maintain user trust in automated decision-making.
• System Responsiveness: Optimize inference latency to ensure AI features (such as chatbots or automated routing) meet the Real-Time requirements of logistics operations.
• User-Centric Feedback Loops: Collaborate with cross-functional teams to translate user feedback into technical model improvements and feature refinements.
• Quality Assurance: Proactively debug and resolve production issues that impact the end-user experience, ensuring high system availability and consistency.