Job Title: Data Scientist (AI/ML Architect) – Distributed Systems & MLOps Location: Jersey city, NJ/ Atlanta, GA (Hybrid/ Remote) Job Type: Full-Time | [Contract / Permanent] • Architect and implement distributed training strategies utilizing frameworks like Horovod or DeepSpeed. • Deploy and manage ML models using containerization (Docker, Kubernetes) and serving frameworks (TensorFlow Serving, TorchServe, Seldon Core). • Implement robust model monitoring and drift detection systems. • Leverage MLOps best practices for CI/CD of ML pipelines and models. • Profile and optimize model performance for low-latency inference. • Integrate with various data storage solutions (e.g., distributed file systems, vector databases). • Contribute to the development of internal AI/ML infrastructure and tooling. • Troubleshoot and debug complex distributed AI/ML systems. Key Skills: • Deep understanding of Machine Learning paradigms (Supervised, Unsupervised, Deep Learning, Reinforcement Learning). • Expertise in Python and relevant scientific computing libraries (NumPy, SciPy). • Proficient in deep learning frameworks (TensorFlow, PyTorch) and their ecosystems. • Strong experience with data pipeline orchestration tools (Airflow, Kubeflow). • Expertise in feature engineering platforms (Feast, Tecton). • Solid understanding of distributed computing concepts and frameworks (Spark, Dask). • Experience with containerization and orchestration (Docker, Kubernetes). • Knowledge of ML model serving frameworks (TF Serving, TorchServe, Seldon Core). • Familiarity with model monitoring and drift detection techniques. • Strong understanding of data serialization and storage formats (e.g., Parquet, Avro, Protocol Buffers). Job Summary: We are seeking a highly experienced AI/ML Architect to lead the design and implementation of scalable, distributed, and production-ready ML systems. The ideal candidate will possess deep expertise in machine learning, deep learning, MLOps, and distributed computing, with hands-on experience in both research and engineering. You will architect and build the infrastructure required to train, deploy, and monitor high-performance ML models at scale. Key Responsibilities: • Architect and implement distributed training strategies using frameworks like Horovod, DeepSpeed, or Ray Train. • Design, deploy, and manage ML model serving with frameworks such as TensorFlow Serving, TorchServe, and Seldon Core. • Utilize Docker and Kubernetes for containerization and orchestration of scalable AI workloads. • Establish CI/CD pipelines for ML systems using MLOps practices and tools (e.g., Kubeflow, MLflow, Airflow). • Implement model monitoring, data drift detection, and performance alerting systems in production. • Optimize model inference pipelines for low-latency, high-throughput serving, including model quantization or hardware-specific tuning. • Integrate with distributed data systems (e.g., S3, HDFS, Delta Lake) and vector databases (e.g., Pinecone, FAISS, Weaviate). • Build and extend internal AI/ML tooling and platforms for experimentation and reproducibility. • Debug and resolve complex issues in distributed AI/ML workflows, both at the infrastructure and algorithmic level. Required Skills & Qualifications: • 8+ years of experience in Machine Learning Engineering or AI Infrastructure roles. • Deep understanding of Supervised, Unsupervised, Deep Learning, and Reinforcement Learning paradigms. • Expert proficiency in Python and scientific computing libraries (e.g., NumPy, SciPy). • Hands-on experience with TensorFlow, PyTorch, and associated ecosystems. • Strong experience with distributed training frameworks (Horovod, DeepSpeed, Ray, Dask). • Proficient in containerization and orchestration tools (Docker, Kubernetes). • Familiarity with ML model serving (TF Serving, TorchServe, Seldon Core). • Solid experience with data orchestration tools such as Apache Airflow, Kubeflow Pipelines. • Understanding of feature stores (e.g., Feast, Tecton). • Knowledge of serialization formats like Parquet, Avro, Protocol Buffers. • Strong analytical and debugging skills for AI/ML pipeline optimization. Preferred Qualifications: • Master’s or PhD in Computer Science, Machine Learning, or related field. • Experience with LLMs, Gen-AI, or multimodal models. • Familiarity with multi-cloud AI deployments (AWS/GCP/Azure). • Exposure to AIOps, observability tools, and real-time feature pipelines. • Knowledge of GPU acceleration, multi-node training, and inference optimization (ONNX, TensorRT). Soft Skills: • Strong leadership and architectural thinking • Excellent collaboration with cross-functional teams (data engineering, DevOps, product) • Effective communicator, both technically and strategically