Weekly Review — July 20-26, 2026
written by Stefan Christoph
This is the Weekly Review, a Sunday digest of everything that went up on the blog this week, plus a short list of things I read but didn’t write about. If you only have ten minutes on a Sunday, this is the one to read.
Prefer to watch? This narrated walkthrough covers the whole digest in a couple of minutes.
This Week on the Blog
Hyrum’s Law: Who Pays for a Broken API in the Age of Agents?
Hyrum’s Law says that with enough users, every observable behavior of your API becomes something someone depends on — documented or not, the implementation quietly becomes the interface. The 2026 wrinkle is that APIs are increasingly consumed by AI agents through layers like MCP, and an agent can dynamically route around a breaking change by writing its own glue, so the user may never notice the break. But that flexibility is not free: the workaround runs per call, so you pay in latency and spend, and each bespoke patch is non-deterministic in a way a versioned library never was. Agents trade a loud, fixable failure for a soft, recurring tax — which is why API discipline matters more now, not less. Second in “Laws & Disorder.”
The Quadratic Wall: Beating O(n²) Attention
Standard self-attention costs on the order of n² in the sequence length, in both arithmetic and the attention-score memory it materializes, and almost every long-context architecture headline of the last few years is an attempt to beat that wall. The honest way to read them is by which cost axis each one moves: FlashAttention keeps exact O(n²) arithmetic but cuts materialized memory from O(n²) to O(n); sparse attention, linear attention, and state-space models like Mamba lower the arithmetic exponent itself. Those axes don’t move together, so no single technique simply wins — and the oldest takeaway still holds: the cheapest quadratic is the one you never trigger, so context discipline and retrieval can beat brute-forcing a longer prompt.
Stop Guessing Which Coding Agent to Use — Benchmark It
Generic coding-agent leaderboards don’t tell you which model, wrapped in which CLI, does your kind of work at a price you’re happy to pay, so I built a small task set mirroring my real work and ran an open-source framework that measures cost, quality, and latency together. Two findings surprised me: the priciest model (Opus 4.8) missed a task that a mid-priced one (Sonnet 4.6) passed at a third of the cost, and when comparing the same model across two tools, the per-credit rate you actually pay can flip which tool looks cheaper. All numbers are from real runs I did on my own tasks.
The Thread This Week
All three posts refuse the surface answer and count the real cost. Hyrum’s Law is about the contract you didn’t write but still pay for — now paid per call by an agent’s silent workaround. The Quadratic Wall is about the exponent hiding inside “just put the whole document in the prompt.” And the coding-agent benchmark is about the leaderboard that was never measured on your workload. In each case the move is the same: look past the headline and measure the thing that actually bills you.
Further Reading
Things I read recently that didn’t get their own post, all public:
- Transformers: The Definitive Guide (O’Reilly) — Nicole Koenigstein’s end-to-end guide to transformer internals: attention, encoder/decoder, training, and multimodal. The book-length companion if the Quadratic Wall post left you wanting the mechanics underneath attention.
- Test coverage is a vanity metric (Markus Bestehorn) — Coverage percentage measures the wrong thing; meaningful testing validates behavior, edge cases, and regressions. A good frame right after a post about measuring what actually matters instead of a vanity number.
- A community effort to standardize MCP server testing (Angie Jones) — As MCP becomes the agent-to-tool protocol, testing standards become critical for enterprise reliability. Directly relevant to the Hyrum’s Law worry about agents depending on undocumented behavior.
- 37 videos from “Code with Claude” (Prasad Rao) — A free library of sessions on Claude Code workflows and production agent patterns. Useful viewing if the coding-agent benchmark made you want to go deeper on the tooling.
- Practitioner lessons from building AI agents (Prasad Rao) — Patterns, gotchas, and architectural decisions from hands-on agent building. Field notes to pair with the benchmark’s “measure it yourself” message.
- A multi-agent strategy team with an adversarial critic panel (Joost de Leij) — A managing-partner agent orchestrates 21 associates that generate rival strategies and run them past a critic panel, in about fifteen minutes. An interesting take on evaluating outputs adversarially rather than trusting a single pass.
Until Next Sunday
That’s the week: a law that explains why your “non-breaking” change broke someone, an algorithms race to bend attention’s quadratic, and a benchmark that beats a leaderboard because it runs on your own work. The thread is refusing the surface answer and counting the cost that actually shows up on the bill. What’s a cost in your own stack you’ve been assuming instead of measuring?
About the Author
Stefan Christoph is a Principal Solutions Architect at AWS, focused on agentic AI, media & entertainment, and helping builders move from demo to production. He writes about AI architecture, developer productivity, and the future of software.
This is a personal blog. Opinions expressed here are my own and do not represent the views or positions of my employer.
❤️ Created with the support of AI (Kiro)