Reading List – 2026-11-03
I’m struck by how much an AI agent’s usefulness depends on boundaries outside the model: per-call policy enforcement, predictable usage limits, and the cost of routine work. Those constraints shape what teams can safely automate and afford to run. Separately, the Cloud Run example shows why persistent files don’t necessarily provide database-safe writes. Why Agents Are Soon Only Going to Have 1 Tool Code mode moves tool orchestration from repeated model turns into a script that can chain dependent calls and filter results before returning them. In the article’s example, a script scans selected pull request files and sends only…
Keep reading
Reading List – 2026-11-02
The context an agent receives shapes both the code it produces and the changes it can review. I’m interested in the tension between preserving architectural decisions and compressing large diffs to fit a review budget. Separately, rising DDR4 prices complicate the calculation around reusing memory: savings can come with narrower motherboard choices. Here's how much RAM cost in 2020 vs. 2026 Older DDR4 isn't insulated from the memory price squeeze. A 32GB DDR4-3200 kit listed for $94.99 in 2020, while the cheapest comparable listing in October 2026 was $198; these are retail snapshots, not annual averages. What catches my attention…
Keep reading
Reading List – 2026-10-30
I’m thinking about what it takes to put agents to work responsibly. Fewer tokens per task don’t settle the question of production cost, and an agent’s authority needs to be clear before it acts. Separately, remembering Patrick Stewart brings me back to Picard’s mix of listening and decisive leadership. Mistral Large 4 (Le Chonk) is now Open Weight Token efficiency matters alongside benchmark scores when coding agents make many model calls. Mistral reports that Large 4 scored 48% on the Artificial Analysis Coding Agent Index, roughly matching GLM-5.3 while using 47% fewer tokens per task. What catches my attention is…
Keep reading
Reading List – 2026-10-29
AI comparisons get more interesting when the unit is a completed task, not just a token or a benchmark score. I’m struck by how much can sit outside the headline: security evaluations without guardrails, and agent status updates that consume tokens. Those details change what a promising result might mean in production. Gemini 4 Carbon Carbon’s external security benchmarks were evaluated without guardrails, which matters when comparing the results with a deployed workflow. The announcement reports 88.2% on SEC-bench Pro for finding vulnerabilities and 54.8% on ExploitGym for validating exploitability. I’d keep that evaluation detail in mind when judging how…
Keep reading
Reading List – 2026-10-28
Several pieces point to the same engineering problem: passing the first check doesn’t guarantee a safe outcome. A model can propose a valid business change, but the orchestrator still needs to recheck live records and control writes. An authorized SSH session can also exceed its intended permissions. I find that distinction useful: correctness has to hold across the whole path, not just at the initial gate. The Counting House The Counting House puts lower-tier models on hardened rails inside human workflows, then measures their performance against human counterparts. That’s a different design problem from Wheelhouse, where agents can see the…
Keep reading
Reading List – 2026-10-27
Several pieces turn on where agent autonomy meets a boundary: what an app exposes and what an assistant can read. Another asks what tools can execute when a model’s trace says otherwise. What catches my attention is how much trust still sits in permissions and enforcement, not the agent’s account of its intent. Elsewhere, the tradeoffs are more operational: latency, memory, and the cost of human review. Apple is adding an AI chatbot to Pages, Keynote, and Numbers Apple is opening Pages, Keynote, and Numbers to third-party agents through MCP on macOS 27 or later, alongside its own context-aware assistant.…
Keep reading
Reading List – 2026-10-26
Across the agent pieces, autonomy keeps landing on familiar engineering boundaries: who can write, what a worktree isolates, and what context survives a reset. The sources describe controls, not proof that they hold under every failure mode. I find that distinction worth keeping in view: delegation can reduce hands-on work, but it still leaves teams owning authorization, coordination, and review. Partyline: A group voice application for Reticulum, built on LXST Partyline's slow-link rooms trade latency for fewer, fuller packets: they pack audio to fill the link's MTU and give listeners larger jitter buffers to smooth bursty arrivals. The project recommends…
Keep reading
Tokens Over Hires
I priced every Claude Code session on my laptop since May and it came to about fourteen thousand dollars. Our whole bill, for both founders and every pipeline, is about twenty-six thousand a year. That’s less than one engineer, and it isn’t the constraint. The constraint is how many agents two people can supervise without splitting their attention, and that changes what a hire is for.
Keep reading
Reading List – 2026-10-23
The agent stories show several ways to constrain or inspect access, from keeping credentials out of an agent’s hands to narrowing network calls and previewing tool requests. They leave a harder boundary open: what can the agent do once access is granted? I keep coming back to that gap. Elsewhere, Jev’s schema-valid output makes a separate point: valid structure isn’t evidence of a sound decision. 1Password wants AI agents to handle logins without ever taking a peek at your passwords 1Password’s agent model keeps the credential out of the agent’s hands: after you approve a request, 1Password fills it directly…
Keep reading
Reading List – 2026-10-22
Delegated authority, change validation, and current design constraints all become harder to take for granted as software does more work on its own. A purchase approval can bound what an agent may do; as code volume rises, the notes suggest CI capacity matters for validation, while stale design docs can misdirect later work. I’m left wondering how teams keep those checks effective when oversight is intermittent. While Amazon blocks AI agents, Walmart rolls out the welcome mat Walmart is treating agent purchases as a distinct authority to manage, not just another checkout path. Its Agent Profiles separate an agent’s transactions…
Keep reading
Reading List – 2026-10-21
The agent workflows here depend on written constraints and design docs staying coherent as code changes. That shifts maintenance work from prompt setup to keeping those inputs and checks current. Separately, rising inference prices make token-heavy features an operating commitment. I’d treat caching and workload-specific cost checks as design concerns, not cleanup for later. New prize launches for books co-written with AI The Prix Pascal makes AI co-writing explicit: entrants submit unpublished French fiction, 10,000–50,000 words, and identify the model and prompt. I like the sound of it; it even has me considering whether to dust off my half-finished manuscript,…
Keep reading
Reading List – 2026-10-20
When AI can shape what people believe and take actions on devices or business phone lines, reliability stops being only a model-quality question. These notes point to two different failure surfaces: weak sources can gain authority through citations, while delegated actions can impose real costs. I’d keep asking where verification and operational boundaries belong. Publish fake stats, get cited in LLMs LLM retrieval can reward a statistic for matching a query, not for being true: the experiment’s fabricated numbers were fetched and later cited in research. The author says systems retrieve a few pages under latency and cost limits, then…
Keep reading
Reading List – 2026-10-19
Automation earns trust only when its evidence holds beyond the setup that produced it. Harness² refines its harness without ground-truth feedback, Timestral’s benchmark leaves production drift unresolved, and Phare says it will refine detection over the alarm’s lifetime. I keep coming back to the validation burden: better behavior matters, but systems that adapt need a credible way to show what changed and whether it remains safe. Personal software factories are a waste of time The case against a personal software factory is about opportunity cost: build with chat tools instead of building a system first. I’m not fully persuaded; building…
Keep reading
Reading List – 2026-10-16
AI agents can reach beyond the person who opted in, and code-review agents can find the right file without finding the bug. Those are different failure modes, but both make me wary of treating access or activity as evidence of safe outcomes. The useful engineering question is where to keep human judgment, and what controls make automation failures detectable and reversible. iPhone Duo orders are already delayed into 2027 The Duo's online delivery estimates slipped to late January within an hour, though the article says high demand, limited supply, or both could explain it. The Mac Studio I checked yesterday…
Keep reading
Reading List – 2026-10-15
Experimental model capabilities are easier to expose than to trust: explicit API flags keep the risk visible, while clean-room pilots can hide production conditions. Faster vulnerability discovery still leaves people to do the remediation. I find the gap between what a system can demonstrate and what a team can safely absorb worth keeping in view. Fancy Curl Sites Full-blog curl output was straightforward because Gemroff already emitted ANSI-marked plaintext, and the static site generator supported modular output formats. A wrapper template and a few generator changes exposed every post; Nginx selected that output for curl user agents, with an override…
Keep reading
Reading List – 2026-10-14
As agents move from producing suggestions to running work, the engineering question shifts from output quality to authority: what can they change, and how do we know when to stop them? Executable checks can surface failures; scoped credentials and approval gates bound what agents may do. I’m interested in whether those controls stay legible across workflows that span several systems. OpenAI, Google pushing new AI models easily tricked into making nudes, researchers say RISE’s uncomfortable finding is that a downloadable Qwen attacker and Gemini judge could bypass image safeguards about 13 percent of the time. Its useful engineering detail is…
Keep reading
The Questions Are the Job
Agents took the half of engineering we used to hand to new grads. What’s left is the half that was always hard to teach, and a conversation with a student made me try to say what it is.
Keep reading
Reading List – 2026-10-13
Capability is easier to demonstrate than sustain. Across the AI pieces here, the pressure shows up in recovering from unfamiliar hardware, keeping review capacity ahead of automated output, and updating defenses as abuse tactics change. I find that a more useful lens than raw performance: what matters is whether the operating loop can keep pace with the system it supports. AI models may be capable of self-replicating without access to a datacenter The difficult part isn’t raw compute so much as standing up inference on unfamiliar hardware: package choices and machine-specific debugging are serial, and failures are hard to recover…
Keep reading
Reading List – 2026-10-12
Across these pieces, capability keeps running into the conditions around it: review capacity for generated code, a harness that shapes an agent’s results, and clear boundaries for connected tools. I’m struck by how often the hard part shifts from getting a model to act to making its behavior inspectable and dependable in the workflow. That’s where deployment judgment still matters. AI Is a Tool. Using It Well Takes Skill. An LLM can be a tool even when it automates work: a programmer can ask it to explain unfamiliar code, check the answer against the source, test a suggested change, then…
Keep reading
Reading List – 2026-10-10
Today’s reading connects the tools around AI-assisted development with the judgment behind using them: security review, documentation, token costs, and programming as expression. Anthropic launches free AI security scans for open-source projects OSS Scanner trades human triage for faster, more frequent scans: its vulnerability reports are fully model-generated, and Anthropic warns they may be incorrect or invalid. That speed matters only if project maintainers can sort useful findings from noise; the article notes that some projects already struggle with AI-generated bug reports. What catches my attention is whether scan frequency can help without adding more triage work to already-stretched maintainers.…
Keep reading
Something went wrong. Please refresh the page and/or try again.