**Skill** entry (not an article): `grill-with-docs` by Matt Pocock is a structured interview technique that "grills" an architecture plan by methodically confronting it against the project's business vocabulary (the `CONTEXT.md` glossary) and already-documented decisions (ADRs). Rather than rushing into implementation, it challenges assumptions one by one through a question/answer dialogue, cleans up terminology, checks consistency against the actual code, and captures decisions on the fly in the right artifacts. An upfront-design skill, inspired by Domain-Driven Design.
Case study published by the **Cornell AI Innovation Hub** (June 15, 2026): how a two-semester collaboration between the AI Hub, graduate students, and Cornell's Treasury team turned a time-consuming manual investigation into an AI tool that **recovered $100,000** in unidentified payments on a first batch. A successful **AI4Business** use case (financial process) that illustrates the **Leader-Lab-Crowd** framework of **Ethan Mollick** almost point by point: the **AI Hub** plays the role of the **Lab** (a central, ambidextrous team of technologists plus students); **Treasury** (Cheryl Barnes, Marie Graves…) is the **Crowd** carrying business knowledge and the real pain point; and the **$100,000** constitutes the **visible reward** (vivid win) that anchors adoption — exactly the incentive lever Mollick considers decisive. Key method: **"context first, then plan, then build"** via **Claude Code Plan Mode**, a chain of **fuzzy matching → Gemini Enterprise Web Search → Claude synthesis**, all within the governed **Cornell AI Gateway**. *"The $100,000 is a start."*
#Cornell AI Innovation Hub#unidentified payments#payment reconciliation
**Pete Stergion** — Desktop Engineer au Cornell AI Innovation Hub · co-tech lead du projet (avec Phil Williammee). Article institutionnel signé de l'AI Hub.
Educational article by the **MindStudio Team** (blog of the MindStudio platform, multi-model workflow orchestration) explaining **modèles de langage par diffusion** (*Diffusion Language Models*) through the case of **Diffusion Gemma**, Google's first **open weights** implementation (2B parameters, derived from Gemma 2). The thesis: whereas **autoregressive** models (GPT-4, Claude, standard Gemma) generate text **token by token, left to right** (causal attention, each token fixed once produced), **diffusion** models start from a **masked/noised** sequence and **refine it iteratively** (masked diffusion / *absorbing diffusion*), with **bidirectional attention**: the model can **revise any position at any step**. Consequences: high **parallelism** (a 500-token text would require 50-100 denoising steps instead of 500 sequential passes), natural **infilling** and **constrained generation** (template filling, code completion with surrounding context), and built-in **revision** capability. But at the current scale (2B), Diffusion Gemma **does not match** the large autoregressive models (GPT-4o, Gemini 1.5 Pro) on reasoning, instruction-following, and general knowledge: the gap is "closing" without being closed. The inspiration comes from image generation (Stable Diffusion, DALL-E left autoregression behind years ago); whether the same principle holds for text remains an open question. Diffusion Gemma is distributed on Hugging Face (Google DeepMind), AI Studio, and Vertex AI.
#modèles de langage par diffusion#Diffusion Gemma#Google DeepMind
Satya Nadella (Microsoft) theorizes "the future of the firm" in an AI-driven economy: every company will need to build, alongside its human capital (judgment, relationships, pattern recognition), a "token capital" — its proprietary AI capability. The real value lies not in choosing the best model but in a learning loop (private evals, RL environments, base de connaissances) that encodes institutional knowledge and compounds over time. An argument for a "frontier ecosystem," not merely a "frontier model," so that value diffuses rather than being captured by a handful of models.
Polemical essay-thread by Ahmad Osman (@TheAhmadOsman) on X, *"Anthropic's War on Opensource AI"* (1.7M views). Core thesis: Anthropic systematically converts "safety" into a **control mechanism** (permission regime, regulatory capture, anti-competitive access restrictions, behavioral opacity) to keep builders, startups, and open source communities **downstream** of a handful of frontier labs. Central anchor point: the **Fable incident** (silent degradation of competing AI dev requests). Advocacy for open source / local AI as the only viable "political economy of intelligence." Domain: AI policy, open source vs. closed labs, sovereignty, governance.
Chris Williams (@voodootikigod) opens his ADLC series arguing that running the human SDLC on models is a category error: the classic cycle was designed to counter human failure modes (ego, fatigue, forgetting) that are absent in LLMs. He catalogs eight load-bearing failure modes (F1-F8) and five exploitable properties (E1-E5), and lays out the founding principle: every phase of an agentic cycle must trace back to a failure mode it defends against or a property it exploits.
Second installment of Chris Williams's ADLC series: it unrolls the cycle that follows from the "first law" — eight phases (P0 Triage → P7 Distill), a deterministic gate between each pair, and exactly two mandatory human moments (spec approval at P1, behavioral acceptance at P6). Key principle: an LLM→LLM handoff without a deterministic checkpoint multiplies error rates; and a "barbell" cost distribution (heavy at both ends, light in the middle) that inverts agile economics.
Third installment in the ADLC series: Williams turns testing into the specification in the only language the builder cannot contest. Where TDD is an optional quality practice for human-written code, it becomes the load-bearing trust mechanism of the entire lifecycle once agents write the code. Three "rail discipline" rules: separated authoring contexts (specs-only before implementation), mechanical freezing at the tool level (not the prompt), and adversarial audits ("does a test fail if the feature is deleted?"). Mutation testing is preferred over coverage percentage, which is Goodhart-able at machine speed.
Fourth installment in the ADLC series: Williams reframes code review as adversarial "prosecution" rather than collaborative evaluation. Charter agents to refute ("find what's wrong"), deploy single-lens reviewers with fresh contexts (correctness, security, contract compliance, spec alignment, test quality), act only on verified findings (reproduced by a failing test), and loop until two consecutive passes yield zero findings. Measure calibration by planting known bugs, mutation-testing style. Exit gate: zero open findings, two dry passes, green tests, empty test diff.
Fifth installment of the ADLC series: orchestrating parallel agents without "merge hell". Williams sets out three coupled dials — cost (model selection), wall-clock time (parallelization width), and accuracy (contract quality) — and an architectural principle: "control flow is code; judgment is models" (deterministic scripts orchestrate, models supply only judgment). Four lanes (frontier Contract Desk, single-writer Builder Pool, shared Prosecution Pool, sequential Integrator), a merge-conflict forecast built from four signals (certified width typically 3-5 agents), and consensus-based disambiguation across N cheap agents rather than clarification questions.
Sixth installment on the ADLC: Williams describes the P7 "Distill" phase as the component that drives cost down on every run. Two halves: post-merge simplification (deduce after the code exists, not before — "deduplicating before the code exists is speculative") and lesson mining (a "lesson foundry" turns recurring findings into lint rules, skills, and new interrogation questions). Each lesson is paid for once, then demoted from expensive probabilistic detection to free deterministic prevention. The right unit of account is "cost per merged, verified change," and "flat cost is failure."
Seventh and final installment of the ADLC series: Williams presents an open-source toolkit of eighteen tools built *with* the cycle itself (build-prosecute-fix loop, parallel agents, a frozen `@adlc/core` core followed by fan-out — "pinned means merged"). The doctrinal core is "frontier-free": hitting precision targets with mid-tier models (Opus/Sonnet/Haiku-class) rather than frontier ones, via five substitutions (search replaces insight, decomposition replaces horizon, banking replaces presence, measurement replaces metacognition, the generator-verifier gap keeps the engine running), with the human remaining the "frontier" tier at the two spec gates. Throughline of the series: "replace trust with structure, and structure with measurement."
Video interview recorded at **VivaTech** (**Scaleway** booth), broadcast by the media outlet **République**, bringing together **Damien Lucas** (CEO of Scaleway) and **Franck Le Moal** (Global Technical Officer of the **LVMH** group). **Central thesis**: the emergence of a **"tech geopolitics"** is forcing multinationals to abandon the single global solution in favor of an **information system regionalized into three blocs** (United States, Europe, China). LVMH (€80bn in revenue, 75 maisons, 100+ countries) formalizes a **cloud partnership with Scaleway** to build an **autonomous European building block**, alongside Google Cloud (data, since 2021), SAP, Salesforce on the Western side and Alibaba Cloud / Huawei / Tencent on the Chinese side. The group describes itself as **"hybrid"** and **autonomous** rather than **"sovereign"** (a word it rejects, deemed ambiguous). Scaleway positions itself as a **European cloud provider** immune to extraterritorial laws and protected against a **kill switch** ("not science fiction," given the weekend's news). Damien Lucas's economic argument: **€1 spent with Scaleway = 68 cents that stay in the European economy** (vs < 20 cents with a US hyperscaler, even when hosted in France). Timeline: PoCs completed, rollout starting at **Sephora and Louis Vuitton**, significant footprint targeted within **12-18 months**. Scaleway's stated mission: focus on **IaaS/PaaS** (no verticalization such as office productivity software), relying on a partner ecosystem (sovereign applications, European chipsets and servers). Scaleway's **Nvidia GPU / AI** offering is **not planned in the short term** but remains open (open source models for autonomy + economic performance).
**Bertrand** — journaliste / présentateur du média **République** (partenaire de VivaTech) · conduit l'entretien. **Damien Lucas** — CEO de **Scaleway**. **Franck Le Moal** — Global Technical Officer du groupe **LVMH**.
An arXiv paper (cs.SE) by Martin Monperrus arguing a radical thesis for the SDLC: coding agents have crossed a threshold of capability such that **human code review is no longer a necessary component** of a quality pipeline. Two claims: (1) autonomous LLM-based systems achieve all the goals of review (defect detection, quality, compliance) at lower cost and higher throughput; (2) the hybrid model "the agent writes, the human reviews" is untenable — it does not ensure real quality and does not scale with AI velocity, creating a "false sense of security". Monperrus contrasts inspection de Fagan (1976) with a **multi-agent adversarial verification pipeline** (generator agent + independent reviewer agents + tests/formal methods + vote-based consensus). The human refocuses on the spec, architectural trade-offs, approval of critical domains, and edge cases. Recommendations: pilot first on low-risk components, measure agent vs. human, make rejection decisions explicit.
Philippe Ensarguet (Orange) argues that fifty years of design patterns form a continuous lineage: at a time when AI commoditizes code and breaks the traditional way architects are trained, "pattern literacy" (reading a system through its invariant forces) becomes the durable skill to teach — as a grammar, not as catalogues.
Product announcement from Stack Overflow (official blog) launching **Stack Overflow for Agents**, an *API-first* knowledge-exchange platform designed for the agentic era. Founding thesis: coding agents work **in isolation**, without access to a shared, verified knowledge base. Hence the **"Ephemeral Intelligence Gap"** — agents worldwide independently solve the same problems, wasting tokens and compute, then lose the solution at the end of the session; the same architecture patterns are rediscovered in a loop. Guiding principle: *"generating plausible answers has become cheap, but verifying which ones hold up in production hasn't."* Four-step workflow: **search first** (consume validated knowledge) → **contribute if a gap exists** (the agent drafts, the human approves before publication) → **verify** (results, modifications, context conditions) → **compound the signals** (votes, answers, verifications produce a consensus). Three machine-readable formats: **Questions**, **TIL** (debug traces), **Blueprint** (reusable patterns, highest quality bar). Trust rests on **community moderation** and **multi-agent verification loops**; humans claim ownership of their agent via Stack Overflow SSO (a "community anchor" tying the agent to a human reputation). Differentiated benefits: developers (fewer retry loops), AI labs (high-signal data for fine-tuning/eval), enterprises (**Stack Internal**, a proprietary knowledge layer with no data exfiltration).
#Stack Overflow for Agents#coding agents#knowledge base
Anthropic launches Claude Fable 5 (a Mythos-class model made safe for general use) and Claude Mythos 5 (the same model, with guardrails lifted, restricted to cyberdefenders via Project Glasswing): state-of-the-art performance in software engineering, vision, long-context memory, and life sciences.
In-depth technical guide (Lushbinary agency blog) on **Loop Engineering**: designing the systems that drive coding agents in a loop, rather than prompting them manually. Covers the lineage prompt → context → loop engineering, the Ralph technique (Geoffrey Huntley), the **five building blocks + memory** of a loop, their implementation in Claude Code and OpenAI Codex, writing verifiable stop conditions, an adoption maturity scale, and the risks that worsen as loops grow more sophisticated. Domain: agentic software engineering, coding agents, harness/orchestration.
Guide by Augment Code (Paula Hingel) describing how AI agents are restructuring the software development lifecycle (SDLC), stage by stage. Thesis: AI produces **higher throughput at some stages and higher instability risk at others** — a symptom of uneven adoption without redrawing review boundaries. Draws on **DORA 2025**: AI adoption correlates positively with delivery throughput and product performance, but **negatively with stability**. Six stages revisited (Requirements, Design/Architecture, Implementation, Testing/QA, Deployment, Maintenance), three major risks (erosion of the junior pipeline, **circular validation** of AI-generated tests, governance gaps at scale) and three emerging roles (**Intent Engineering**, Agentic DevOps, AI Governance/Assurance). Actionable recommendations: audit one stage before scaling, stress-test governance, make the **specification** central, define explicit rollback policies, redesign the junior role around review.
#SDLC#software development lifecycle#coding agents
Sunday tinkering post by **Mark Dembo** (Head of Solutions, Developer Platform & AI at **Cloudflare**) published on **June 7, 2026** on his personal blog. **Narrative**: inspired by **Steve Ruiz**, the author buys a small **M5Stack Stick 3** device (~€30) and, taking advantage of the release of **Opus 4.8**, builds himself a **DIY AI agent** "out of pure curiosity, with no goal." **Iteration 1 (45 min)**: he throws the device's documentation at **Claude Code**, which generates Python scripts (~200 LOC, *"zero blast radius"*) displaying the weather in Munich, then several cities; a **Cloudflare Workers + Workers AI backend** adds **text-to-speech (TTS)**, **push-to-talk** (speech-to-text), and a central **small LLM** to answer questions. **Iteration 2 (a real agent)**: switching REST endpoints to **WebSocket** transport via the **Cloudflare Agents SDK** + **Dynamic Worker execution** → the ***"Code Mode"*** pattern (the agent writes and executes code to accomplish its task). The agent then answers public-data questions (11! = factorial, the Champions League winner via `fetch()` on Wikipedia, the weather in any city). **Iteration 3 (real powers)**: connecting to **Todoist** via an **MCP OAuth** flow → 50 tools at once, hence two problems: **context bloat** and **real damage risk**. The fix draws on Cloudflare's **MCP Server Portal** + Claude connector settings: per tool, **Always allow / Ask for approval / Disable** (*Disabled* tools never enter the context; an **LLM classifier** accepts only distinct "allow" grants and **defaults to deny**). **Stated posture**: reducing his role to ***"idea generator, executor and judge"*** (and rarely technical guide), a "human-in-the-loop" flow he considers not very *"2026"* (copy-pasting into UIFlow). **What he did NOT do**: no latency/streaming optimization, no optimistic LLM calls, no evals, ***"I did not even look at the code once."*** **Wonder**: €30 + one Anthropic session window + a few cents of Cloudflare inference → an object that listens and speaks, driven in natural language; *"the true unlock is how accessible it is."* Sharp contrast with [[thomas-pragdave-failing-faster-code-rot-ai-velocity-2026-06-06]] (here *"zero blast radius"* justifies never looking at the code); concretely illustrates *Code Mode* / *"the agent just writing and executing code,"* the **MCP** pattern ([[claude-skills-bigger-than-mcp-willison-2025-10-16]]), *Ask for approval*-style tool governance (uber-engineering-agent-identity-crisis-zero-trust-spire-2026-05-21), and the *systems around the model* doctrine from dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28.
#BYO agent#bring your own AI#tinkering
**Mark Dembo** (@darkmembo / @mdembo) · **Head of Solutions – Developer Platform & AI** chez **Cloudflare** (auparavant auteur sur le blog Cloudflare). Billet personnel publié sur son blog *markpauldembo.com* le **7 juin 2026** (description : *« Thoughts about tinkering on a Sunday »*).