Essays, interviews, and lecture transcripts from the faculty and alumni. The thinking behind the curriculum, in long form.
We publish what we'd want to read on a Sunday morning — work that earns the reader's attention by being honest about how hard it actually is.
Issue XII · Lead
14 min · 6,200 words
№ 012
Most engineers learn to write code long before they learn to read it. We teach the reverse — and after twelve cohorts, the difference shows up everywhere: in code review, in onboarding, in what people choose to build next.
Read the essayA staff engineer from a payments company on the patterns that survive contact with reality. Less about syntax than about what to leave alone.
↗Three of our faculty re-read the EWDs and respond — section by section, generously and irritatedly, as Dijkstra would have wanted.
↗A working bibliography from our type-theory faculty — the foundational papers that still earn their place after a decade of re-reads.
↗Three signatures of a too-early refactor, and the question we ask in code review before promoting any helper to a module.
↗Edited transcript from week 3 of Database Internals. On why B-trees beat hash tables for the cases databases actually face — and what it costs when you forget.
↗Not because the language is timeless, but because the practice of reading and writing trees is. A short defense of the four-week detour.
↗What changes when reviews are read aloud, and what happens when authors stop apologizing in PR descriptions. Field notes from three cohorts.
↗Edited from week 5 of Systems & Performance. Why a 64-byte boundary is the most important number you've never thought about — and how to feel it.
↗Hannah Lindqvist, Marcus Okonkwo, and Saoirse Whelan — three Lattice graduates who started companies — on what carried over from the curriculum and what didn't.
↗Twelve papers, books, and lectures we recommend before applying. Read at least three. We'll know.
No tracking pixels. No links you didn't ask for. Just the new pieces, when there are new pieces.
— By submitting you agree we'll only send when there's something to send.