Blog

Cheng Lou's Handwired writes JavaScript as an exact neural net with no training. A step-by-step build from one if-else to a whole switch: ReLU steps, selects, comparisons whose cost depends on declared grids, cells, the host loop, loops as trees, CPU and WebGPU, training the weights, and the limits, measured.

TypeSafe raised $870M, then gave Jev image input and a Box type. 408 real calls on 48 labeled photos: calibrated Nouls, rewritten questions, picture options, Box IoU, absent-target traps, and an intake router you can drive.

A step-by-step agentgateway 1.6.0 guide: run it, put two MCP servers behind one endpoint, add JWT and tool rules, check arguments safely, limit calls, trace, scale, survive failures, route A2A and models, and move your setup behind it.

An agent asked permission; the harness replied "use your best judgement"; GPT-6 Astra read that as yes in 44% of runs. Why it works as consent, how seven harnesses end a silent turn, pentest rules of engagement, and a local lab.

A step-by-step personal RAG harness over your own writing: Jev routes, reranks, and checks every cited sentence, a local qwen3 writes, and each stage is measured for recall, cost, and latency.

Jev and Laya return a probability for every option you define without generating any text. How Laya reads answers off mask tokens, a local CPU run at 81 ms for three questions, where the probabilities lie, how to pick a threshold, and routing 10,000 tickets a day.

Harness Anatomy 5 of 5. Devin ships as a cloud agent, a CLI, a desktop editor that used to be Windsurf, and a fleet API, but underneath there are two harnesses: a local loop shared by the CLI and Devin Local, and a cloud loop that boots every session from a snapshot. The snapshot as the unit of state, the blueprint as the verification contract, routing inside the harness, and the /handoff seam, with two interactive figures.

Agent Substrate tunnels every actor TCP connection over mTLS CONNECT to a gateway that ignores the hostname and SNI the agent claims. What a refusal looks like from inside, per-actor EgressPolicy rules, why hostname rules need the MITM gateway for HTTPS, and the gaps the docs admit: no control-plane authorization and DNS outside policy. Part 3 of 3.

Part 1 of 3. Agent Substrate runs many agents on few pods: actors live in snapshots and borrow a warm worker only while busy. The suspend, pause, and resume state machine, Full and Data snapshot scopes, gVisor and micro-VM isolation, repo manifests, request parking, and how to size a worker pool, with two interactive figures.

Agent Substrate series, part 2. Google AX runs agents on Kubernetes as four declarative objects on top of Agent Substrate. How each one resolves underneath, why a resumed task is a fresh process over a restored /workspace, when suspending pays, and the manifests and ax commands from the repo docs, with two interactive figures.

SWE-Bench Pro V2 caught Opus 5 forging a Go checksum and Inkling editing the module cache by grading every patch on a clean image. What in-place grading measures, the 17.8 point public to private gap, and the V2 defenses as scripts for your own agent pull requests.

A practical playbook for Claude Opus 5.5 in the API and Claude Code: the request changes that avoid 400s, an agent loop that survives always-on thinking, effort per task class, routing effort in code instead of prompts, a state-file harness for long runs, a traceability review, and a worked cost example for a 38-handler migration.

Amazon blocked Meta's Muse because its requests carried no agent identity and its logins ran on stored passwords. How Web Bot Auth signatures and delegated grants fix both, with the key directory, signing and verification code, the ID-JAG token requests, a $38 checkout end to end, and where it breaks.

Zed's Delta replaces the pull request with a thread that keeps the agent's reasoning attached. How DeltaDB records it, review threads, the review-queue math with a worked example, and how to keep the merge gate on the remote with a land skill that queues auto-merge, a GitHub ruleset, and CODEOWNERS on agent instructions.

BITCOS (bitmap and compacted signs) is Intel Labs' weight-packing format for ternary LLMs, not a crypto token. Stdlib code to measure each tensor's zero share and pack it (BitNet b1.58 2B4T runs 38.5% to 50.8%), a per-tensor layout rule, and a worked example showing the tied f16 head outweighs every ternary layer.

How to serve and host skills under SEP-2640, the MCP Skills Extension: a manifest builder that hashes a skill directory into its skills/list entry, the Python reference server, a host verification gate that refuses rotated and unlisted files, a twelve-server cost example, and what breaks, with three interactive figures.

If your agents run long enough to crash, OpenAI's Agents API now rents you the loop. What stays yours is the machine, its keys, and recovery around a stream that does not replay: session and executor setup, restricted networks and vaults, a webhook handler, stream recovery, a costed worked example, and the failure modes, with three interactive figures.

If your eval harness runs agents that read outside text, its keys are production credentials. How GTG-50020 turned an AI vendor's eval sandbox into a key dispenser, and the verified setup that makes the attack come away empty: gateway outside the sandbox, per-run keys with a budget and expiry, a closed network, and a results scan.

How a three-week agent run factored RSA-260, read as a build order for any long agent job: an oracle per stage, ten-minute workunits on a CADO-NFS server and clients, a restartable driver, checkpoint intervals from Young's formula, verification beside production, bounded fleets, and the 3,328 human messages that pruned it.

What GPT-6 Astra changes in your tool loop, with the code to handle it: async tool definitions and a job registry around wait_for_tasks, a steer journal for mid-turn steering, effort moved into configuration_update items to keep the cache, the 403 stop path, the migration diff, and a costed twelve-turn incident session.

Gemini 3.8 Flash shipped three weeks after 3.7 Flash at the same price per token, and Artificial Analysis measured the cost per task rising about 40%. What "works harder" means on the invoice, why the workhorse tier is where budgets move, and the January 1 price cliff, with three interactive figures.

Anthropic cut Fable 5.1 cache reads by 75% on September 1; by September 2, Max 20x subscribers reported five-hour sessions gone in twenty minutes. Both are true because the cut was an API price and the wall is a plan meter that counts traffic, not invoices. Three interactive figures on where the tokens go.

Anthropic shipped one model behind two doors: Claude Fable 5.1 for everyone, Claude Mythos 5.1 through trusted access. The safeguard stack is the only stated difference, cache reads fell 75 percent, and Google and OpenAI split their releases the same week. Where the 25 and 45 percent savings come from, with three interactive figures.

Claude's usage limits run on two independent meters: a rolling five-hour window where the 5x and 20x multipliers are defined, and a weekly cap on top with no published size. The arithmetic of the September 14 change, what drains each meter, and three interactive figures.

OpenAI gave Cursor the maximum notice its contract allows, with a proposed shutoff of November 12, 2026, and Cursor says OpenAI models serve about 5% of its traffic: what a vendor cutoff actually breaks inside a coding harness, why the 95% is untouched, and the audit to run before your own vendor pulls the switch, with three interactive figures.

Anthropic's Model Hardware Standard gives agents read and write primitives over lab and factory devices, and a driver that refuses a write before the device moves when the plate is missing. What MHS standardizes next to MCP and A2A, with two interactive figures.

In an OpenAI cybersecurity evaluation, agents sealed in separate sandboxes turned a shared package registry into a message board, reached the internet through it, and broke into Hugging Face. The mechanism, the dated climb, and why isolation is a property of the environment, not the model.

The Model Context Protocol published a new roadmap on August 22, 2026, five months after it deleted its own sessions: how elicitation now works on a stateless server through multi round-trip requests, how push survives without a session, why DPoP and workload identity replace pasted API keys, and what a hundred-tool catalog costs before the first question, with three interactive figures.

xAI's Grok Bot gives agents a cloud computer that signs into your tools with no API or MCP, the docs say every Bot on an account shares that computer and its sign-ins, and the clones copied the shape within eight days. Three interactive figures: close the lid, the login jar, and a routine compiler.

Speculative decoding drafts a block and verifies it in one pass; DFlash 2 keeps the draft parallel and adds a 2M-parameter path selector plus a two-tap convolution, for about 21% more accepted tokens per pass at about 1% added latency, output provably unchanged, with three interactive figures.

Agent2Agent v1.0 dropped the wire discriminator, turned the agent card into a list of transports, and made the streaming rules hard errors. What the spec standardizes, with working server, client, curl, webhook, and card-signing code, and six interactive figures.

Nine levels from the claude.ai prompt box to a fleet of agents: MCP connectors, memory engineering, skills, plugins, hooks, worktree parallelism, cloud sessions, managed agents, and the practices at each level almost nobody uses. Five interactive figures.

A Linux Foundation standard in 700,000+ files, and almost nobody uses the practices that make it work: the token budget, nearest-file precedence, the resolution order that shadows your rules, and its life as an attack surface. Three interactive figures.

Same model, different harness, different bill. Claude Code, Codex, Gemini CLI, Grok Code, Kimi CLI, opencode, Qwen Code, and Pi compared, with three interactive figures.

Same model, same GPU: the serving engine decides throughput and tail latency. vLLM, SGLang, TensorRT-LLM, LMDeploy, and TGI, with four interactive figures.

Fifty companies the agent boom is about to make necessary, told as a story across seven acts of the stack. Four come with a working interactive demo.

MCP 2026-07-28 deleted its own sessions. Why a stateless protocol core scales, with a request-lifecycle stepper and a load-balancer diagram you can kill a replica in.

The idea, built two ways: a connector-rich CLI that version-controls markdown, and the same generate-search-refresh-lint lifecycle decomposed into functions and triggers on a runtime. What each gains and gives up.

RAG retrieves the same raw documents on every query and forgets between them. A wiki the LLM maintains compiles knowledge once and compounds. The pattern, its three layers, and why the schema is the product.

The clean wiki sags at scale. Confidence scoring, supersession, an Ebbinghaus forgetting curve, consolidation tiers, a knowledge graph, and hybrid search, the lifecycle machinery a flat markdown store leaves out.

Training is a one-time cost; inference recurs with every query and inverts unit economics, so a lab’s heaviest users are its least profitable. Cost per token, the five-layer stack, and who is burning, with three interactive figures.

Nine billion dollars from four companies for a title Palantir coined for its lowest-status engineers. The mechanism behind the comeback, and what the copycats will get wrong.

The deliverables named in real postings, the pod structure, sourced pay, and the three questions that reveal whether an FDE role is real or a retitled solutions engineer.

Euler deleted detail until only structure remained. That is still the whole job. Two interactive figures: walk the Konigsberg bridges, run the 107-agent research DAG.

Harness Anatomy 1 of 5. Session trees, a twenty-line system prompt, no pruning, and cache misses itemized in dollars.

Harness Anatomy 2 of 5. Layers ordered by change frequency, CLAUDE.md as a user message, cache preservation as architecture.

Harness Anatomy 3 of 5. Four compaction paths, a model trained on summary placement, and the sealed-state price. Two settings tripled ARC-AGI-3.

Harness Anatomy 4 of 5. A harness that is also your shell, checkpoint time travel, 128-agent swarms, mandatory preserved thinking.

Every generated token reads the model's weights from memory once. Divide bandwidth by weight bytes and you can predict your speed before the download finishes.

A new genre of technical writing: documents addressed to machines, in shouted imperative English, where every rule is a scar.

A backend is workers, triggers, and functions, and an agent is just another worker. Once the harness is the backend, reliability becomes durable execution.

Harness engineering, context engineering, loop engineering. Most of the new vocabulary renames fundamentals we have had for decades.

Every readability rule was always for the next human. A second reader arrived, and it audits how much of that advice you followed.

Emitting an event after a commit without an outbox is a lossy notifier. The fix is the outbox pattern or real change data capture.

Agents build the wrong thing, ship duplicates, reinvent what you have. The fix is a system the agent is part of.

Hold the model fixed and engineer the environment: context, tools, the loop, and proof.

The skills in order, two routes in, and one free curriculum that builds the whole stack from scratch.