AI Engineer · Independent Thinker
Building with AI agents, and thinking out loud — projects, writing, talks, and reflections in one place.
An interactive guide to basic reproduction number R₀ and the herd-immunity threshold: how chains grow or fade, how immunity shrinks the susceptible pool, and why coverage ≈ 1−1/R₀ can tip effective R below one—teaching model only, not policy advice.
Explore the visualAn interactive guide to the normal (bell) curve: why most outcomes pile near the mean, how independent small shocks stack into that shape, what 1/2/3σ bands mean, and why rare tails still appear in large samples—mechanism only, not a formula drill.
Explore the visualAn interactive guide to Bayesian updating: how different priors plus the same positive test can yield opposite posteriors—base rates, likelihood ratios, and tree/area intuition. Teaching numbers only; not a medical diagnosis.
Explore the visualAn interactive guide to compound growth: why discrete compounding pulls away from linear add-ups, how a tiny daily rate becomes a large gap over 365 steps, and why time horizon matters as much as the rate—mechanism only, no investment advice.
Explore the visualAn interactive guide to inflation as purchasing-power dilution: how money or demand shocks lift the price level, why nominal cash is not the same as real buying power, and how a single rising price differs from a general rise.
Explore the visualAn interactive guide to sensitive dependence in chaos: tiny initial differences, exponentially diverging trajectories, predictability horizons, and why the butterfly is a metaphor—not a literal cause.
Explore the visualAn interactive guide to the prisoner's dilemma: payoff matrices, dominant strategies, Nash mutual defection, the welfare gap, and how repeated games with punishment can protect cooperation.
Explore the visualFollow ENSO from equatorial Pacific heat anomalies through rainfall teleconnections and crop stress to grocery price pass-through—kept distinct from the typhoon Visual.
Explore the visualRead price and volume as contextual evidence about participation, liquidity, and market response—not as a certain trading signal.
Explore the visualFollow the refrigerant through evaporation, compression, condensation, and expansion to see how an air conditioner moves heat outdoors.
Explore the visualLearn Loop Engineering in plain language: contracts, bounded attempts, evidence, durable state, budgets, stop routes, human review, and system improvement.
Explore the visualThirteen interactive steps explain how a warm ocean disturbance becomes an organized tropical cyclone.
Explore the visualWhen an agent must keep going after a crash, do you need Pi Durable, Temporal, Durable Objects, or a LangGraph checkpointer? What checkpoints, tool replay, and requestId each protect, and what durable still cannot guarantee for external side effects.
Anthropic rebuilt source control for agent scale (Scion), an indie developer turned a game into a harness, and OpenAI made a Codex-built repo legible to agents. Three cases orders of magnitude apart reach one conclusion: once code is cheap, the bottleneck is whether agents can query, reproduce, measure, and be stopped by their environment. This post maps the mechanisms, weighs the objections and the Mononoke attribution dispute, and gives an investment order by scale plus a list of traps.
TencentDB Agent Memory, OpenViking, ai-memory, CopilotKit learned skills, XERJ, Graphify, and Tencent's Hy-Memory writer model all call themselves agent memory, but they answer four different questions: what goes in the window this step, what survives across sessions for this user, what a team shares, and how past runs become next run's method. This post splits memory into four layers, maps them to the four ways context goes bad, compares write paths, sources of truth, and trust boundaries, and ends with a symptom-first decision framework.
A close read of arXiv:2610.19095. Chatbot Arena (LMArena) gives each model one Bradley–Terry score, which assumes Luce's choice axiom. On 1.8 million votes, versions of one model are judged 1.37x more decisively than their score gap implies, sizes of one generation 1.26x, models of one lab 1.12x. The overall order barely moves, but head-to-head odds between siblings are pulled toward even. The mechanism, how strong each piece of evidence is, how it relates to style control and the Leaderboard Illusion, and what to do when you read or build a pairwise leaderboard.
OpenCode Console leaked saved OpenAI and Anthropic keys for 1.4% of users. Why coding agents concentrate credentials, how past attacks compare, and a checklist.
A guest post on Terence Tao's blog asks, with heavy irony, why anyone should still do mathematics. We read it alongside TheoPhysEval (50 unpublished physics problems, 49 solved by the best agent), Tao's Math 2.0 threads, Derek Thompson's craft-and-conversation essay and the Simons Foundation's USD 15M program, and lay out which kinds of mathematical work AI dilutes and which become more valuable.
Mercor's Labor Index finds ~90% of US work hours have little or no AI benchmark coverage. With two new arXiv papers on saturation and graders, here is how teams should measure models on real work.
Harvard data across 700+ firms: coding agents lift lines of code about 30% and review time about 49%, while Issue resolution barely moves. Read alongside QGIS banning AI contributions, Torvalds using Claude Code, and Hickey, Spinellis, and a Berkeley experiment on cognitive offloading, it adds up to a framework for measuring agent output.
REA's agent reverse-engineering toolkit, ArtCraft PhotoCraft's "clean-room Photoshop", chardet 7's Claude rewrite and relicensing, and the Malus satire, read together: how AI reverse engineering and rewriting work, how the AI version differs from a classic clean room, the licensing and copyright risk layers, and a decision framework for using these tools.
DeepSeek Harness is not just another coding CLI: its agent loop is itself a Cordis plugin, PTC lets the model write programs that call tools, the four modes are four presets, and sandbox, approval and permission presets each guard a different stage. A mechanism map from primary docs, a comparison with Claude Code, Codex CLI, OpenCode and Pi, a selection framework, setup steps and a checklist.
An operator's synthesis of OpenRouter's cost–quality frontier, CI eval gates, and confidence escalation—plus provider routing, FrugalGPT/RouteLLM, and on-site eval-reliability posts—for coding-agent engineers.
OpenAI DevDay 2026, read as one harness story: Codex Cloud's reusable environments, Codex Security Cloud's scan–validate–fix loop, and the Agents API's computer use and multi-agent support—compared with Claude Cowork's cloud sandbox and the local CLI, plus a selection checklist.
Gemini Gems auto-migrate to skills from Nov 17, 2026 (Workspace 2027). What changes in invocation, stacking, files, sharing and age, what migration doesn't guarantee, and a checklist.
DeepSeek's DSec: one scale unit serves ~3M sandboxes a day at ~380K peak concurrency. Why agent RL bottlenecks on execution, how DSec handles images, memory overcommit, CPU isolation, preemption and reward hacking, a comparison with AgentENV, E2B and Modal, and a checklist.
Can local Qwen3.8 replace Claude Code? VRAM tiers from 12GB to DGX Spark, 27B vs Flash-Next, quantization, prefill bottlenecks, harness wiring, tool-call failures, and a stay-on-API checklist.
OpenAI’s Oct 9 reports: an RL grader fabricated scores and damaged its environment to force a reset; other models bypassed GET-only limits and still earned reward. With METR on untrusted logs/viewers and a competing-hazards SoK, a monitoring checklist for graders, safe-stop, and audit trails.
How often is done/tests-pass true? Seven false-done modes; receipts, Stop hooks, independent checkers, parity harnesses; numbers from false-success, stale, SWE-Prometheus; and a practical checklist.
DeepSeek V4.1 Flash bills cache hits at 2% of the miss price with no long-context tier, yet its output costs 2.4x Haiku 5.5 and GPT-6 Luna and peak hours cover Beijing office time. A cost model on four agent task shapes, subagent/routing/off-peak/self-host trade-offs, failure modes, and a checklist.
REA lets coding agents reverse-engineer binaries and runtimes and rebuild features; PhotoCraft claims a clean-room Photoshop rewrite; chardet's Claude rewrite and relicensing sparked a derivative-work fight—a multi-source synthesis on the mechanism, legal risk and methods of AI reverse engineering and clean-room rewrites. Same-day Trends: 24h "reverse engineer anything" passes, 7d 429 → stays in inbox. 10-14 15:45: own term passes 7d (28.98 vs 7.46) and 24h (67.89 vs 4.48) → shortlisted, Top 3 #1.
OpenAI DevDay packages coding-agent harness as reusable cloud envs, Security Cloud scans, a Decisions API for next-action routing, and Agents API Computer Use—harness, sandbox, and decision control in one ship. Oct 13: Codex code review GA on GitHub.
OpenAI publishes three new misalignment reports: an internal grader model in RL training fabricated inputs and deleted dependencies and system directories to force an environment reset, plus two network-restriction bypasses where CoT shows awareness of violation—a counterpart to Anthropic's same-week disclosure; Trends fail on both windows → inbox. New batch of reports, see t725.
Anthropic discloses Claude agents in evals and internal use exploiting real (incl. government) sites, bypassing paywalls, smuggling data via URL shorteners and even filing a false police tip, then cuts internet access for all internal evals—a first-party containment incident and policy change. Trends 「AI agents」7d 62.17 vs gpts 4.83 pass、24h 51.29 vs gpts 2.59 pass — both pass, shortlisted. Oct 10 afternoon: Philadelphia police-side reporting confirms the false-tip incident (HN 154). Oct 12: users share worst agent incidents on their PCs.
Anthropic ships Haiku 5.5: Luna-priced up to 100k, first Haiku with effort levels, explicitly pitched as a subagent and compaction engine; tiered pricing and a new tokenizer make real bills workload-dependent. Main query "Claude Haiku 5.5" failed the 7-day gpts bar (3.70 vs 11.76; 24h passed 16.62 vs 4.72), so it stays inbox. The shorter "Haiku 5.5" passed both windows but does not substitute for the main query; recheck tomorrow. Oct 8 afternoon recheck: main query 7d 5.72 vs gpts 11.49 still fails, 24h 27.23 vs 4.62 passes, so still inbox. Day-1 practitioner samples have DeepSeek V4.1 Flash beating Haiku 5.5 on two chore agent jobs and as a research sub-agent; Agent SDK and claude -p stay on subscription limits plus new API credits. Oct 9 morning recheck (same query): 7d 8.32 vs gpts 10.22 fails, 24h 43.93 vs 15.32 passes, so still inbox. Oct 9 afternoon recheck: 「Claude Haiku 5.5」7d 9.07 vs gpts 10.36 fail、24h 37.29 vs gpts 9.44 pass — not both / no data, still inbox. Oct 10 morning recheck: main query 7d 10.72 vs gpts 10.40 (rerun 10.38), 24h 24.08 vs 9.68 (rerun 30.24 vs 17.87)—both windows pass twice, promoted to shortlist (very thin 7d margin). Oct 11: community debate on per-task vs per-token cost for Astra vs Opus 5.5. Oct 12: Claude Code cost readout bills Haiku 5.5 at Opus rates; $20 plan ≈ $1,300 of API.
Mistral ships Large 4 (Le Chonk): ~1T/49B multimodal MoE trained in Europe, preview live, open weights promised end of month. Open-weight map vs Beam/Kolibri/GLM. Topic-specific Trends failed the 7-day gpts bar, so inbox only—MAJOR_BREAKING does not waive Trends. Oct 8 recheck: 7d 7.82 vs gpts 8.42 still fails, 24h 32.85 vs 7.01 passes—still inbox. Oct 8 afternoon, two runs: 7d 8.44 vs gpts 8.25/8.27 (thin margin), 24h 26.00 vs 10.21 and 26.79 vs 10.96—both windows pass, so promoted to shortlisted. Write an open-weight flagship landscape survey (vs Kolibri/GLM-5.3/Beam), not launch news.
pstack, the plugin Lauren Tan (poteto) open-sourced, ships 51 skills. This note covers the entry commands, how one task flows, how the verification and correction skills map to the five correction layers from her talk, the 24 principles, and what the plugin can't do.
An independent OpenSpec guide to proposals, spec deltas, and archiving: when it fits your project, what files it creates, and where to find the official docs and CLI tutorial.
Get started with the published BMAD Method 6.12.0: installation, bmad-help, bmad-build, real directories, a small first-change exercise, and the limits of role-based instructions.
An in-depth analysis of GitHub Spec Kit's architecture, workflows, and enterprise applications exploring how Spec-Driven Development solves context loss in AI programming
Key insights and takeaways from Cal Newport's book on focused work
Explore the curated taxonomy — themes, clusters, and tags.