Topic
Everything filed under Engineering, newest first.
RSS · JSON · All topics
An Aggregate Is a Function, Not a Sum
Alexander Novikov explains why treating a streaming report aggregate as a function of its source documents, rather than a running sum of inserts, makes duplicates and late-arriving documents in a Kafka-to-ClickHouse pipeline stop needing manual corrections.
17 min · 3,851 words
Introducing Kinference: An Open Source Inference Control Plane for vLLM and Kubernetes
vLLM and collaborators introduce Kinference, an open-source inference control plane for Kubernetes that adds token rate limiting, multi-tenant controls, and operational primitives on top of vLLM serving.
4 min · 1,034 words
LLM Judges for Identity Resolution at 4x Lower Cost
We wanted to test whether LLMs and decision models, specifically Jev and Luna, could solve a hard problem that we encounter regularly (grading identity resolution results). We found that while both models performed okay out of the box (88.8–97.3% accuracy), implementing a hybrid model that combines Jev + Opus achieves frontier accuracy at > 4x lower costs. - Luna performs better than Jev out of the box, but the gap disappears on harder cases. - Jev's confidence is better calibrated. Most of its mistakes come with low confidence its right. Most of Luna's mistakes come with…
11 min · 2,468 words
Tuist Runners: Going down to the bare metal on macOS and Linux
Managed macOS and Linux CI runners with colocated caching and insights down to the metal. Invite-only while we grow capacity.
10 min · 2,231 words
From Static to Polymorphic Inference: A new era for production open-source inference systems
Models, traffic and chips are constantly changing. Polymorphic Inference adapts to each, and Together Compiler uses AI agents to keep production inference tuned.
7 min · 1,647 words
How We Optimized Browser Use for Deep Agents
LangChain explains how it made browser use efficient for Managed Deep Agents: context-tuned Stagehand tools, selective observation, and a human handoff path so agents can browse without drowning in page noise.
8 min · 1,820 words
Accelerating engineering for the physical world
OpenAI describes how Astra and Codex help design, simulate, and manufacture physical products—from an Astra-designed Codex device to collaborations with Chip Ganassi Racing and Mazda.
5 min · 1,203 words
A rebuild could clear 28% of our client's container CVEs. The rest needed evidence.
A red container-vulnerability test in the middle of a SOC 2 observation window. A rebuild could clear 28% of the findings. The rest took scanners that never agreed, VEX claims customers act on, risk acceptances with an owner and a date, and a reachability test that proved my own argument wrong.
14 min · 3,246 words
New context engineering challenges in late 2026
Harrison Chase surveys late-2026 context engineering challenges for agent builders: longer horizons, tool noise, memory tradeoffs, and why harness design now matters as much as the underlying model.
4 min · 892 words
Making Kubernetes AI-Ready for NVIDIA Vera Rubin on Day One
Configuring and optimizing Kubernetes for new GPU hardware such as NVIDIA Vera Rubin is complex and error-prone. Drivers, kernels, runtimes, networking, operators, and schedulers must work together, with settings tuned to the hardware and workload. Small mismatches can prevent workloads from running or limit performance. The NVIDIA AI Cluster Runtime (AICR)—part of the NVIDIA DSX AI Factory Platform—simplifies this work by capturing tuned configurations in reusable recipes, generating bundles for existing deployment tools, and validating the running cluster. It will help…
12 min · 2,706 words
Altera launches Agilex 2X, a four-family FPGA/SoC portfolio for Digital and Physical AI, with Agilex 27 first and a unified QuartusIQ agentic design flow.
3 min · 615 words
Can GROUP BY still be optimized after all these years?
Pivot walks through how analytical databases execute GROUP BY and shares a novel optimization for GROUP BY queries with a Top-K operator above them, pruning partitions during execution for a 1.6x speedup on a ClickBench-style query.
6 min · 1,372 words
RWE-bench: Real World Engineering Bench
Xplore launches RWE-bench, a benchmark of 20 expert-designed mechanical, electrical and aerospace engineering tasks (crankshaft CAD, DDR3 PCB routing, turbine-vane CFD) graded by deterministic physics-based verifiers with no LLM judge. Across 11 frontier models, GPT-6 Astra leads at a 37% average pass rate, Claude Opus 5.5 scores 29% and GPT-6.1 Sol 25% at about $10 a run; five models score 0%.
5 min · 1,145 words
How Vercel Agent accesses production safely
Vercel explains how Vercel Agent is allowed to touch production safely—scoped permissions, approval gates, auditability, and the boundaries that keep automated agents from becoming an uncontrolled production risk.
8 min · 1,825 words
Animations for System Builders: Metastability and Lost Writes
Marc Brooker shows how LLM-built animations backed by real test traces help build intuition for distributed-systems failures, including how a Raft cluster can lose writes when one node's storage lies, and metastable failure.
2 min · 509 words
SBD: We Have a Filesystem at Home
Fly.io's Daniel Botha introduces the Sprites Block Device (SBD), a simpler object-storage-backed block device that replaces the clever custom filesystem Sprites launched with, explaining why dumber checkpoints and 'just a disk' made Sprites more reliable.
6 min · 1,282 words
Whatever Happened to Fire Escapes?
Practical Engineering traces the history of the external iron fire escape in American cities, why it became the default retrofit after early fire egress laws, the problems that made it unreliable, and how building codes shifted to protected interior stairwells and other egress designs, with the tradeoffs regulation creates.
10 min · 2,407 words
What requirements don't you have?
Sean Goedecke argues that elegant software design comes from identifying requirements you can drop, such as fresh data, concurrent writes, or low tail latency, citing Unicorn, S3, StatsD, SQLite, and Cursor's git backend, and warns that unchecked consistency demands petrify systems.
5 min · 1,096 words
Evolving the cloud-native foundation for the next generation of AI
Microsoft Azure describes how it is advancing cloud-native infrastructure so organizations can build, run, and optimize applications and agents at scale for the next generation of AI workloads.
7 min · 1,552 words
It’s always DNS, so why are we running our own?
S2 engineer Cristian Caciuc recounts why the streaming-storage company replaced AWS Route53 and now runs its own authoritative DNS server so customers' new basins resolve instantly, and how it has held up in production.
6 min · 1,489 words