ML Model Training & Validation

This course teaches you how to evaluate and design custom neural network architectures for real machine-learning tasks. You start by learning how to compare common model families—such as CNNs, RNNs, and Transformers—and match them to task needs, data patterns, and compute limits. You then learn how to construct custom architectures using layers, activations, and regularization techniques that improve generalization and training stability. Through videos, readings, hands-on practice, and guided coach support, you build models in PyTorch and test how design choices affect performance. By the end of the course, you can confidently select topologies, justify architectural decisions, and design models ready for real-world deployment.

This short course helps you deploy and optimize scalable machine learning workloads in the cloud using managed AI services. You’ll start by learning how distributed training jobs work on platforms like Amazon SageMaker. Then you’ll configure training pipelines using Spot Instances and autoscaling features, gaining hands-on experience with real-world deployment patterns. Finally, you’ll dig into monitoring and optimization: reading GPU utilization logs, exploring CloudWatch metrics, and making recommendations that balance performance and cost. By the end, you will know how to right-size an ML workload, select efficient instance families, and justify architecture changes based on data.

This intermediate course teaches you how to design scalable, reliable AI systems that work in real-world production environments. You’ll learn how to build end-to-end architectures that meet throughput, latency, and fault-tolerance goals, and you’ll move from conceptual design to detailed component diagrams and interface specifications. Using industry patterns adopted by modern ML teams, you’ll practice estimating QPS, defining autoscaling rules for the inference layer, structuring data flow between the feature store and model API, and instrumenting your system with a monitoring stack. By the end of the course, you will have created a complete architecture document—including diagrams and interface definitions—that engineering teams can use to implement a scalable AI product.

"Integrate and Optimize AI Services Seamlessly" is an applied, intermediate-level course designed for engineers and ML practitioners who want to build reliable, production-ready AI systems. Across focused, hands-on lessons, the course explores how real-world services communicate using APIs, message queues, and structured serialization formats. Learners gain practical experience integrating prediction services with gRPC and protobuf, improving consistency, performance, and cross-language compatibility. The course also guides participants through deployment health essentials, including interpreting Prometheus metrics, spotting early warning signs during canary releases, and making safe decisions to stabilize or roll back new versions. Through real scenarios, interactive activities, and expert-led demos, students develop the confidence to ship AI services that are fast, resilient, and operationally sound in modern distributed environments.

This short, hands-on course helps learners adapt and optimize deep learning models for real-world use. Learners begin by exploring how transfer learning accelerates model development when data is limited. Through guided practice, they fine-tune a pretrained model, adjust freezing and unfreezing strategies, and troubleshoot common training challenges. The course then shifts to evaluating model configurations for deployment, focusing on accuracy, latency, memory footprint, and efficiency. Learners experiment with optimization methods such as hyperparameter tuning and quantization, compare multiple model setups, and make evidence-based recommendations for production environments. By the end, learners can confidently balance accuracy and performance constraints to choose the right model for their needs.

This course helps learners transform scattered AI preprocessing code into clean, reusable, and testable Python utilities that meet modern MLOps expectations. Across two focused lessons, learners explore advanced programming constructs—such as generators, decorators, and structured logging—that make ML workflows modular and maintainable. They then apply software-engineering principles to design standards-compliant Python packages that integrate smoothly into real AI pipelines. Through videos, readings, hands-on exercises, and a guided Coursera Lab, learners practice refactoring preprocessing steps, structuring packages using current Python packaging standards, managing dependencies, and writing unit tests with pytest. By the end of the course, learners will have the skills to build and test a functional Python package suitable for internal PyPI publishing and production-ready machine learning work.

This short course helps you build and validate ML-ready data pipelines with confidence. You’ll start by learning how to design ETL workflows that ingest, clean, and partition large datasets using tools like Airflow and Spark. You’ll see how real teams manage click-stream logs, handle nulls, and prepare partitioned training data at scale. Next, you’ll evaluate data quality, governance, and lineage so your pipelines remain trustworthy and reproducible. You’ll work with practical techniques like schema drift checks, expectations suites, and audit-ready lineage records. Through short videos, applied readings, hands-on practice, and a final graded assessment, you’ll walk away knowing how to engineer reliable pipelines and validate them for production use.

Machine learning systems shift over time, making structured testing essential. In this short course, you’ll learn how to evaluate ML pipelines using unit, integration, and smoke tests and how to detect data drift across critical features. You will also create automated regression test suites that compare new model outputs to golden datasets, helping you catch degradation early and deploy reliably. Through concise videos, readings, hands-on practice, and guided coaching, you’ll define meaningful ML test cases and configure nightly pytest suites. By the end, you will have a practical, reusable testing framework you can apply directly to real-world ML pipelines.

This short course helps you build and optimize machine learning workflows using TensorFlow 2.x. You’ll start by structuring an end-to-end pipeline that includes data ingestion with tf.data, model definition with Keras, and custom training with checkpointing for reliability. You’ll then learn how to optimize your models for deployment using TensorFlow Lite, including post-training quantization and latency benchmarking. Along the way, you’ll see how ML engineers measure performance, evaluate tradeoffs, and deploy models to mobile and edge devices. Through hands-on practice and real-world examples, you’ll learn to think like an applied ML practitioner who builds efficient, production-ready TensorFlow systems.

This course helps you break down complex ML systems into clear, reusable parts and communicate them using practical abstractions. You’ll learn how to separate ingestion, feature serving, inference APIs, and monitoring components while creating flowcharts and pseudocode that guide implementation. Using examples such as real-time fraud detection and feature store workflows, you’ll practice decomposing systems and designing abstractions engineers depend on. Through short videos, readings, hands-on practice, a coach-guided reflection, and a 45-minute ungraded lab, you’ll build skills used across ML engineering and MLOps roles. By the end, you’ll be able to confidently analyze ML systems and produce artifacts that support scaling, clarity, and production readiness.

Document AI: Project & API Writing teaches you how to communicate AI systems with clarity, structure, and precision - skills that are essential for ML engineering in real organizations. In this course, you’ll learn to document model architectures, data schemas, training procedures, and evaluation summaries in ways that support onboarding, debugging, and reproducibility. You’ll also create developer-facing API documentation with request and response schemas, examples, error behaviors, and usage notes. Through hands-on practice and a full MkDocs documentation lab, you’ll build a complete, developer-ready documentation site for a prediction API. By the end, you’ll be able to turn raw ML projects into professional, discoverable, and maintainable technical documentation that teams rely on.