--- title: "Foundations" canonical: "https://current.tinyfish.ai/issue/13/foundations" publication: "TinyFish" publisher: "TinyFish" language: "en-US" guide: "https://current.tinyfish.ai/llms.txt" permalink: "https://current.tinyfish.ai/issue/13/foundations" issue: 13 section: "foundations" description: "Conceptual clarity earned from building at scale" authors: ["Rina Takahashi"] published: 2026-01-26 updated: 2026-01-22 articles: 11 --- # TinyFish > CURRENT is a journal about AI agents in production, for the engineering and product leaders who have to make them work. Drawing on experience building enterprise web agents, it explains why the demo succeeded and the deployment didn't, and what that means for choosing vendors, briefing the board, and avoiding mistakes others have already made. [Read on the website]() ## Issue 13 — Foundations Conceptual clarity earned from building at scale ### What Building Teaches Builders - [Why Teams Keep Choosing the Wrong Web Infrastructure]() — By [Rina Takahashi]() · Published 2026-01-22 · Format: Field Guide The production incident starts the same way: a team's web infrastructure hits scale and something fundamental breaks. Not a bug they can patch—an architectural mismatch they can't fix without rebuilding. They chose extraction tools for orchestration work, or automation frameworks for workflows that need reasoning. The system collapses because the foundation was wrong from the start. ### Tools & Techniques - [When You Need to See Exactly What the Browser Saw]() — By [Rina Takahashi]() · Published 2026-01-22 · Format: Tools in Context Authentication fails on fifteen Japanese hotel sites. Succeeds everywhere else. The logs say "session timeout"—but that's not what happened. Something about regional redirects and cookie state went wrong in a way you can't reproduce locally. Traditional debugging assumes reproducibility. Web agents at scale generate contextual events that happen once, under conditions you didn't anticipate. When failures depend on production context you can't control, you need to see exactly what the browser saw. - [When You Need to Understand Patterns Across 1,000 Failures]() — By [Rina Takahashi]() · Published 2026-01-22 · Format: Tools in Context Fifteen authentication failures surface across a thousand concurrent sessions. The operational question: do these represent a systemic pattern—new bot detection, changed authentication flows—or fifteen independent issues that happen to coincide temporally? Traditional observability collects metrics and logs. Manually correlating dashboard data, looking for patterns in noise. At scale, you need infrastructure that recognizes whether failures cluster around specific conditions. Pattern recognition across chaos becomes essential when individual precision isn't enough. ### Pattern Recognition - [Agents Break Authorization's Core Assumption]() — Published 2026-01-22 Four security vendors shipped agent identity solutions this month. CrowdStrike bought SGNL. Microsoft released Entra Agent ID. Qualys and Exabeam launched competing products. They're racing to solve the same problem. ### Questions Worth Asking Building an AI agent feels straightforward until production arrives. Your agent meets real users, real data, real constraints. Things start breaking in ways demos never revealed. We've learned that the questions you ask before deployment predict whether you'll scale successfully or spend months debugging surprises. These aren't theoretical concerns. They're the specific gaps between "works on my machine" and "handles production traffic reliably." The questions below cut through marketing to what actually matters at scale. They're what we ask ourselves when evaluating options, born from watching patterns of what works and what fails. - **What's Your P95 Latency?** — [Source]() — Published 2026-01-22 Tag: Processing Time Five seconds versus forty-five seconds per request. That gap determines whether you meet SLAs or create backlogs. Demos show median performance. Production runs on your worst-case latency under load. That's what breaks customer experience and operational capacity. - **Does It Give Different Answers?** — [Source]() — Published 2026-01-22 Tag: Output Consistency Run the same query three times. High variance means brittleness. Your agent is overly sensitive to minor prompt variations. Production systems need predictable behavior. If answers drift across identical inputs, you have a reliability problem before you scale. - **Can You Track Per-Request Costs?** — [Source]() — Published 2026-01-22 Tag: Cost Visibility LLM calls, vector storage, cloud compute. It adds up fast at scale. If you can't measure cost per interaction now, you won't catch runaway expenses until they've already happened. Cost-blind optimization leads to bloated infrastructure nobody budgeted for. - **What Happens When APIs Fail?** — [Source]() — Published 2026-01-22 Tag: External Dependencies External tools introduce latency and failure modes you don't control. When a third-party API times out or returns errors, does your agent degrade gracefully, retry intelligently, or cascade failures? Tool dependencies require evaluating reliability alongside functionality. - **How Will You Detect Degradation?** — [Source]() — Published 2026-01-22 Tag: Performance Drift Roughly 91% of ML models degrade over time. Seasonal behavior shifts, data distribution changes, environmental factors. Performance erodes silently. Without continuous monitoring and drift detection, you discover problems only after users complain or business metrics drop. - **Does Your LLM Judge Match Humans?** — [Source]() — Published 2026-01-22 Tag: Evaluation Validity Automated evaluation scales beautifully but only if it correlates with human judgment. Teams optimize for LLM-as-judge scores without validating them against real user assessment. Then production performance disappoints. Human evaluation is slow and expensive. It's also the gold standard. ### In Dialogue With Complexity - [An Interview With The GDPR Exemption That Never Exempts Anyone]() — Published 2026-01-22 We met Dipro in a Brussels office that looked more like an evidence locker. Stacks of legal briefs climbed the walls. A whiteboard showed increasingly desperate calculations about notification costs. In the corner, a poster: "Disproportionate Effort: Your Safety Valve*" The asterisk led to microscopic fine print nobody could read. ## Other sections in Issue 13 - [Daily Brief]() - [Market Pulse]() - [Practitioner's Corner]() - [Echoes]() - [Vision]() ## More - [Publication guide]() - [Issue archive]()