--- title: "English Has No Compiler" newsletter: "MLOps Community" date: 2026-07-09 source: https://aaif.live/newsletters/mlopscommunity/2026-07-09-english-has-no-compiler --- # English Has No Compiler *Plus… agent cost control, API access patterns, real-money guardrails, and a 35-tool CLI coding agent report.* *MLOps Community — Agentic AI Foundation, 2026-07-09* As part of the transition to the Agentic AI Foundation, the email address this newsletter comes from will soon change to demetrios@aaif.live To help make sure you keep receiving it, hit reply and tell us the AI buzzword you're hearing most. We'll make a leaderboard from the replies. My vote is going to loop engineering. The new address is one of a few changes you’ll notice, alongside some new podcasts in the mix, including a few hosted by the AAIF's very own Alex Salkever [https://www.linkedin.com/in/alexsalkever/]. ## Out the Window A bigger context window is not the same as memory. It is just a more expensive place to forget things. What matters more for agents: long context or persistent memory? [CONTEXT](https://go.mlops.community/NL_HT1_Jul09) [MEMORY](https://go.mlops.community/NL_HT1_Jul09) ## Access Granted Have to trust the accuracy of the people who accessed the poll to tell us what matters more in production. ## Curated finds to help you stay ahead ## JOB OF THE WEEK ## The Next Programming Language Is English Natural language is becoming a programming interface, but it does not come with the guarantees developers are used to getting from formal languages. * Why AI coding changes the abstraction stack, replacing deterministic translation with probabilistic systems that need new safety nets. * How durable execution gives agents and tools a logical process that can survive client failures, network drops, and infrastructure chaos. * Why MCP tasks matter for long-running work, from streaming analysis to invoice processing, human approvals, durable timers, and scale-to-zero infrastructure. The big shift is a programming model that keeps agent work alive when prompts, networks, tools, and humans fail. [https://podcasts.apple.com/us/podcast/the-next-programming-language-is-english/id1505372978?i=1000775638432](https://podcasts.apple.com/us/podcast/the-next-programming-language-is-english/id1505372978?i=1000775638432) [https://home.mlops.community/home/videos/the-next-programming-language-is-english](https://home.mlops.community/home/videos/the-next-programming-language-is-english) [https://open.spotify.com/episode/67u426zix3nNBhwQqllF7E?si=994866c59d6f4c10](https://open.spotify.com/episode/67u426zix3nNBhwQqllF7E?si=994866c59d6f4c10) ## AI Agents Should Be Treated Like Hackers Your agent does not need every API. It needs the right slice, at the right time, with proof that it should have it. * MCP can connect agents to enterprise systems, but most APIs were built for trusted engineers, not autonomous software. * GraphQL offers a useful pattern for agent access because schemas expose business meaning and support field-level permissions. * The next challenge is richer context, better access patterns, audit logs, and accountability around business decisions. Faster agents need narrower doors and better receipts. [https://podcasts.apple.com/gb/podcast/ai-agents-should-be-treated-like-hackers/id1505372978?i=1000775663953](https://podcasts.apple.com/gb/podcast/ai-agents-should-be-treated-like-hackers/id1505372978?i=1000775663953) [https://www.youtube.com/watch?v=HMh63kmgOKc](https://www.youtube.com/watch?v=HMh63kmgOKc) [https://open.spotify.com/episode/5YARel9OOiIxCp9LEJM7v9?si=jVHPD0zRR66bhvG4XJHlwA](https://open.spotify.com/episode/5YARel9OOiIxCp9LEJM7v9?si=jVHPD0zRR66bhvG4XJHlwA) ## Omnigent: Composition, Control, and Collaboration for AI Agents Your agent may be right, but it can still burn money on the wrong model. * Why agent workflows need portable context, memory, and configuration across models and harnesses. * How forking conversations and agent debates can expose missed questions before work goes sideways. * Why tokenomics needs both developer-level visibility and centralized governance. The payoff is practical control over agent work without pretending one model fits every task. [https://podcasts.apple.com/us/podcast/omnigent-composition-control-and-collaboration-for-ai/id1505372978?i=1000775368328](https://podcasts.apple.com/us/podcast/omnigent-composition-control-and-collaboration-for-ai/id1505372978?i=1000775368328) [https://home.mlops.community/home/videos/omnigent-composition-control-and-collaboration-for-ai-agents](https://home.mlops.community/home/videos/omnigent-composition-control-and-collaboration-for-ai-agents) [https://open.spotify.com/episode/12wy1ZWlnpCZGBUzLEIaxd?si=1a3a21637bd9402e](https://open.spotify.com/episode/12wy1ZWlnpCZGBUzLEIaxd?si=1a3a21637bd9402e) ## IN PERSON EVENTS * Munich [https://luma.com/5ckfr8dm] - July 23 * San Francisco, Voice Agents Forum [https://luma.com/voiceagentssf] - September 16 ## VIRTUAL EVENTS * Coding Agents Lunch & Learn Session [https://home.mlops.community/public/events/coding-agents-lunch-and-learn-session-18-community-show-and-tell-3zh201wnwg?agenda_day=6a304c16e47ccefce6894aa3&agenda_filter_view=stage&agenda_stage=6a304c16e47ccefce6894aa9&agenda_track=6a304c17e47ccefce6894aba&agenda_view=list] - July 10 * A Discussion on Loop Engineering [https://home.mlops.community/public/events/a-discussion-on-loop-engineering-me6r7t97sh?agenda_day=6a44eb367234763b95523e18&agenda_filter_view=stage&agenda_stage=6a44eb367234763b95523e1c&agenda_track=6a44eb377234763b95523e4a&agenda_view=list] - July 16 ## Overconfidence Interval [https://docs.google.com/forms/d/e/1FAIpQLScLyUmV7B_fvmiQni6cF0iIFMEvjrL5fivzUjdFci14G_nQCA/viewform](https://docs.google.com/forms/d/e/1FAIpQLScLyUmV7B_fvmiQni6cF0iIFMEvjrL5fivzUjdFci14G_nQCA/viewform) --- Source: https://aaif.live/newsletters/mlopscommunity/2026-07-09-english-has-no-compiler