architecture

Two things reliably surprise teams new to AWS: the network they must build before anything can talk, and an invoice driven by charges nobody chose deliberately. The two are connected, because moving data is where much of the cost hides. This lesson covers the virtual network primitives, the traffic charges that follow from them, and how to read a bill.

Cloud storage looks like a long product list and is really three physical shapes, object, block and file, each with an access pattern it is built for and one it is bad at. This lesson covers those three, the difference between durability and availability that people conflate, what eleven nines actually means at scale, and how to choose a database by access pattern.

AWS offers several ways to run code, and they are not competitors so much as points on one spectrum trading control for operational relief. This lesson walks that spectrum from virtual machines to serverless functions, what you hand over at each step, the cold-start and state constraints that decide fit, and the honest cases where serverless is the wrong answer.

AWS offers hundreds of services, which makes it look like a catalogue to memorise. It is not. This lesson gives the four structures everything else hangs from: the physical geography of regions and availability zones, the account as a blast-radius boundary, IAM as the one gatekeeper every call passes, and the responsibility line between you and the provider.

A recommender has milliseconds to pick ten items from millions, and no model good enough to rank them all is cheap enough to run on them all. The industry's answer is a funnel: cheap candidate generation cuts millions to hundreds, an expensive ranker orders those hundreds. This lesson builds that architecture, its latency arithmetic, and the multi-source retrieval layer real systems run.

Speech in, text, model, text, speech out is the standard architecture and it discards everything not in the words: emphasis, emotion, hesitation, overlap. End-to-end models keep it by never routing through text, and pay with a token rate roughly 185 times higher and far less training data. This lesson covers the trade, and why interleaving is the pragmatic answer.

Pure recurrent models match transformers on broad language benchmarks and lose on the tasks that need exact retrieval from context. That is a capacity limit, not an engineering gap: a fixed state holds a fixed amount. This lesson covers what the evaluations showed, why a handful of attention layers recovers almost all of it, how Jamba and Samba are laid out, and when this is worth adopting.

The other route starts in control theory. A discretised linear system is a recurrence, and time-invariance turns it into one convolution, which trains in parallel. This lesson follows that line: why HiPPO initialisation matters, why time-invariance is what stops the model choosing what to remember, how selection breaks the convolution, and how the parallel scan gets it back.

Softmax is the one operation forcing attention to materialise an n-by-n matrix. Remove it and associativity lets you rebracket the product so a fixed-size quantity is maintained instead. This lesson derives that step, shows why the result is a recurrent network with a matrix-valued state, explains the two computation modes, and is honest about what the softmax was doing.

Attention costs two separate things people conflate: quadratic compute during training, and a cache that grows without bound during inference. At a million tokens that cache is 344 GB while a recurrent state is 16.8 MB and constant. This lesson separates the two costs, shows why the classical RNN alternative failed, and states the trilemma every architecture since has been negotiating.

Running these tools inside a design practice: documenting that care happened, the professional indemnity question, who owns generated output, coordination liability across consultants, and the parts of the work a client structurally cannot get from a tool.

Documentation is the biggest consumer of hours in a practice and the clearest place tooling helps. This lesson covers specification drafting and why a spec is a legal instrument, submittal and datasheet review, request-for-information handling, and the coordination problem between drawings and specs.

Architecture and engineering divide into design, documentation, coordination and administration, and AI touches them very unevenly. This lesson maps the split, then explains the professional seal: a licensed person certifying their own judgement, which is why the responsibility cannot move to a tool.

Strip any robot to its bones and you find three things: sensors that gather data, a controller that thinks, and actuators that move. Learn how these pieces fit together through the Sense-Plan-Act paradigm, why reactive control sometimes beats planning, and what makes a Roomba and a welding arm both qualify as robots.

The honest case for each. When a monolith is correct, what microservices actually buy you (and what they cost), Conway's law, and how to spot a fake microservices architecture that's actually a distributed monolith.

Explore the data lakehouse, a modern architecture that merges the cost-efficiency and flexibility of data lakes with the performance and reliability of data warehouses. This lesson covers its core components, benefits, and the open-source technologies that make it possible.