Software Engineering for AI
Software Engineering for AI on AI-ML Companion: Write production-quality code that scales and is maintainable. 26 interactive modules with live visualizations, quizzes, and hands-on Python coding.
Start free: Clean Code Principles is fully open to everyone, no account required. The other 25 modules are part of AI-ML Companion Premium; every title and summary is listed below so you can see exactly what the track covers before deciding.
Modules in this track
- Clean Code Principles (free) - Writing readable, maintainable code
- Python Best Practices (premium) - PEP8, type hints, and modern Python
- Project Structure (premium) - Organizing ML projects professionally
- Documentation (premium) - Docstrings, README, and API docs
- Testing with pytest (premium) - Unit tests, fixtures, and assertions
- Testing ML Code (premium) - Testing data pipelines and models
- Testing LLM & Agent Applications (premium) - Snapshot tests, cassettes, and a calibrated judge in CI
- CI/CD for AI Systems (premium) - Pipelines, gates, and rollout for prompts, models, and code
- Mocking & Integration Tests (premium) - Mocking APIs, integration testing
- API Design Principles (premium) - REST, versioning, and best practices
- Design Patterns for ML (premium) - Factory, strategy, and ML-specific patterns
- Code Review Best Practices (premium) - Giving and receiving effective feedback
- Engineering with AI Coding Agents (premium) - Specs, guardrails, and review discipline for generated code
- ML System Design Fundamentals (premium) - Latency budgets, two-stage retrieval, feature stores, safe rollouts
- AI Engineering System Design (premium) - RAG topology, LLM serving, agent runtimes, cost cascades, evals + guardrails
- AI Evaluation Engineering (premium) - Golden sets, component-level scores, calibrated judges, and regression gates
- Inference Engine Internals (premium) - PagedAttention, continuous batching, FlashAttention v3, prefix caching, chunked prefill
- Quantization & Compression (premium) - FP16 to FP8 to INT8 to INT4, AWQ vs GPTQ vs SmoothQuant, FP8 hardware on H100/B200
- Collective Comms & Networking (premium) - NCCL allreduce, NVLink, NVSwitch, InfiniBand, RDMA - what makes multi-GPU serving work
- Hardware Fluency (premium) - H100/H200/B200 vs MI300X vs Trainium2 - HBM bandwidth, FP8 support, MIG economics
- GPU Scheduling on Kubernetes (premium) - KubeRay, gang scheduling, MIG management, Karpenter on GPU spot, topology-aware placement
- Multi-Tenant Routing & SLOs (premium) - Prefix-cache-aware routing, priority queues, SLO admission control, fairness vs head-of-line blocking
- LLM Observability + Cost (premium) - TTFT, ITL/TPOT, KV-cache hit rate, $/1M tokens - the metrics every inference platform needs
- LoRA Serving at Scale (premium) - S-LoRA, Punica - serving thousands of fine-tuned adapters on one base model
- Advanced Decoding (premium) - Speculative decoding, Medusa, EAGLE, constrained output - the latency wins beyond standard primitives
- Cold-Start & Model Loading (premium) - Safetensors, streaming load, Alluxio/Fluid, pre-warm pools - get 90s cold-start down to 5-10s