← Back to Home
Research & Writing

From the Lab

Aug 19, 2026
Research

Past the Peak: Maximum Reasoning Effort Made Our Model Measurably Worse

Our on-premises text seat had run at its chat template's highest reasoning-effort setting since the day it was deployed, because nothing in the stack ever set the value. Measured across 94 benchmark items per setting, that default scored 84.6 on our flat composite against 93.4 at the middle setting, while spending 81 percent more completion tokens; disabling reasoning entirely was worse than both, so the reasoning was doing work and its length was the defect. The dimensions measuring raw reasoning were already at ceiling and could not improve. Everything that moved measures whether the model commits: tool selection, abstention, multi-step execution, grounding. Abstention rose from 66.7 to 100.0, meaning maximum effort had reasoned the model into answering questions it should have refused. Because the settings are inserted sentences rather than compute budgets, we then reproduced the top setting as a hand-written expert-persona prompt and watched tool selection fall to the same 62.5 it scored at maximum effort. We give the method, the non-termination failure at the top of the dial, and an explicit account of what these numbers will not carry.

Read
Aug 11, 2026
Research

The Flat Line: Why a Single Throughput Number Cannot Tell Batching From Queueing

A production inference engine reported the same aggregate throughput at one concurrent request as at sixty-four. It was not batching at all: every request past the first was queueing, and the single benchmark point we had been quoting could not see the difference. Sweeping the concurrency ladder corrected a stored throughput figure by 3.2× and moved a model comparison from a reported ~7× to 17.0×, or 33.9× per GPU, since NVIDIA's Nemotron 3.5 Lightning sustained 5,613 tokens per second across 64 concurrent requests on a single Blackwell card, at sub-three-second p95, and had not saturated when our ladder ran out. We give the diagnostic signature (flat aggregate, linearly climbing p95), the four-configuration sweep that exposed it, a quality result that survived repetition and one that did not, and why a recorded performance number belongs to a configuration rather than to a model.

Read
Aug 7, 2026
Research

Make It Go Red: The False-Green Failure Class in Production Monitoring

For five months a nightly backup reported Complete over a zero-byte archive. An alert rule had been structurally incapable of firing for its entire life. A dashboard rendered missing telemetry as health, and 37 of 49 deployment scripts could silently ship stale monitoring state. We name the failure class, the false green, map its mechanisms in four families from a 24-day production audit, show that verification tooling itself reproduces the class it hunts, and describe the falsification harness now grading all 279 of our alerting controls continuously: registry, structural liveness, mutation-proven firing tests, and end-to-end canaries. Monitoring correctness is a reachability property, not a syntax property.

Read
Jul 21, 2026
Infrastructure

The Day an Agent Ate the Workstation: OOM Forensics of a 178 GiB CLI Memory Leak

An agentic coding CLI ballooned to 178.5 GiB of resident memory during a routine fleet upgrade and froze a production workstation. The kernel killed fourteen innocent processes before touching the culprit. A forensic walk through the OOM killer's victim ordering, why swapless machines turn leaks into freezes, the four-layer guard we deployed the same afternoon, and the upstream report we filed instead of posting screenshots.

Read
Jun 22, 2026
Infrastructure

Power-Resilience by Design: Making a Stateful Kubernetes Fleet Survive Losing Its GPU Nodes

Our GPU compute tier is powered off most of the time by design. That makes any persistent state, or any hard scheduling pin, that lands on it a silent failure waiting to happen. Here is how we audited every hard dependency, relocated all state onto the storage cluster, and then proved resilience the only way that counts: by cutting the power.

Read
Jun 19, 2026
Research

An Immune System for a GPU Cluster: Building a Reflex Arc for Autonomous Self-Healing

We set out to design a self-healing v2 and found the opposite of a greenfield: four generations of self-healing stacked in sediment, maximum architectural sophistication, near-zero closed-loop coverage. The fix was not a fifth system: it was naming the missing shape, and then discovering the real blocker was signal quality, not architecture.

Read
Jun 16, 2026
Research

Markdown Is Sovereign, Postgres Is a Cache: Engineering a Rebuildable Agent-Memory Substrate

An agent's memory is its most dangerous component: poison it quietly and every downstream decision inherits the lie. We built a memory substrate on one governing principle (Markdown is the source of truth, Postgres is a rebuildable cache) and then learned the hard way that our retrieval eval had been measuring the wrong thing all along.

Read
Jun 14, 2026
Research

A First Look at Text-Diffusion LLMs: Benchmarking DiffusionGemma on a Single Blackwell

Google shipped its first open-weights text-diffusion language model: a different generation paradigm that paints a whole canvas of tokens in parallel instead of writing left to right. We benchmarked it within days of release, and a 3.8B-active model beat a 128B production model on hard mathematics.

Read
Jun 8, 2026
Research

Serving a 550B Open Model on Blackwell B200: A Field Recipe and an Honest Agentic Benchmark

Can a 550-billion-parameter open model be served on Blackwell B200 and stand up to a frontier closed model on real agentic tool use? The serving path is a gauntlet of era-specific gotchas, and the most important result in the whole campaign is the one we published, retracted, and corrected.

Read
May 29, 2026
Infrastructure

Right Pool, Right Workload: How Every Database We Ran Was Quietly Paying a 17× Storage Tax

A benign-looking storage warning led to an uncomfortable discovery: every production database on the cluster had its data on an erasure-coded pool, paying a partial-stripe read-modify-write penalty on every single fsync. The disks were fine. The pool was wrong, by about 17×.

Read
May 16, 2026
Research

Characterising a 200GbE RDMA Fabric on Commodity Hardware: From a Hidden PCIe x4 to PAM4 Optics

A routine fabric test found one inter-node link running at 29% of capacity, not broken, just quietly starved of PCIe lanes. Chasing that thread all the way down took us from a hidden slot-bifurcation trap to NCCL collective ceilings to a per-lane optical asymmetry that error-correction was silently hiding.

Read
Apr 30, 2026
Infrastructure

Building One Operator Cockpit for a Sovereign GPU Fleet

A sovereign GPU fleet emits a dozen overlapping signals (GPU box health, Kubernetes workloads, distributed storage, metrics, out-of-band power) and operators were jumping between tools to read them. The engineering story here is what it takes to turn an internal cockpit into something safe to expose: config-driven topology, read-only-by-default gates around every dangerous action, and secret-scanning in CI.

Read
Apr 20, 2026
Research

The Trinity Model Shoot-out: Five Frontier MoE Models on Workstation Blackwell

Five frontier open-weight mixture-of-experts models (MiniMax-M2.7, GLM-5.1, DeepSeek-V3.2, Qwen3-235B and Qwen3.5-397B) served on the same three-Blackwell cluster under llama.cpp TCP RPC and graded on one mechanical 30-prompt suite at matched budget. A hybrid Mamba2+transformer leads at 96.7%, and default thinking modes turn out to be a methodology trap.

Read
Apr 19, 2026
Research

Distributed Inference on Workstation Blackwell, Part 3: A 228-Billion-Parameter MoE Reasoning Model via TCP RPC

Can a single-operator workstation cluster with no NVLink and no InfiniBand serve a 228-billion-parameter mixture-of-experts reasoning model at conversational speed? Part 3 distributes MiniMax-M2.7 at Q8_0, 226 GiB across three Blackwell GPUs, under llama.cpp over plain TCP RPC, and measures the no-optimisation floor.

Read
Apr 18, 2026
Infrastructure

Distributed Inference on Workstation Blackwell, Part 2: Amalgamated System RAM for GPU Dataframe Analytics

Does a workstation-class Blackwell cluster extend beyond language-model inference? Part 2 retargets the same two-node, three-GPU cluster to a full GPU-accelerated click-through-rate pipeline (Parquet ingest, high-cardinality shuffle, target encoding, and multi-GPU XGBoost) on roughly 750 GB of amalgamated system RAM over RAPIDS and Dask-CUDA.

Read
Apr 17, 2026
Research

Distributed Inference on Workstation Blackwell, Part 1: Cluster Bring-Up and Cross-Node Fabric Characterisation

Can frontier open-weight language models be served on workstation-class NVIDIA Blackwell GPUs interconnected by commodity RDMA Ethernet, without NVLink or InfiniBand? Part 1 of an ongoing PureTensor research programme brings up the cluster, characterises the fabric, verifies GPUDirect RDMA, and demonstrates first cross-node inference of Llama 3.1 405B.

Read
Mar 25, 2026
Research

Toward a Rotating Embedded Braneworld Cosmology: Critical Assessment, Formal Viability, and a Research Programme for Geometric Dark Sectors

The third paper in the CMAS research series, and the artefact the methodology was built to produce. Cosmology is the conduit; the question is whether structured adversarial interaction between frontier AI models can produce research-grade technical output on an open problem where no ground truth exists. The paper below is what came out.

Read
Mar 24, 2026
Research

From Intuition to Formalism: Cross-Model Adversarial Synthesis as a Theory Development Tool

Can a vague metaphysical idea, lacking any mathematical grounding, be refined into a falsifiable research programme through structured adversarial interaction between frontier AI models? A case study in speculative cosmology.

Read
Mar 22, 2026
Research

Cross-Model Adversarial Synthesis: Exploiting Latent Space Heterogeneity for Novel Knowledge Generation

Can structured adversarial interaction between independently trained LLMs produce insights that exceed any single model's capability? We present a theoretical framework and initial case study in black hole information theory.

Read
Mar 18, 2026
Research

Agentic Architectures and the Restructuring of Software Engineering Labor

An analysis of how autonomous code-generation agents are compressing multi-month development cycles into days, and what the structural implications are for the software engineering profession.

Read
Mar 14, 2026
Research

Context Engineering: Why Most Enterprise AI Deployments Fail Before They Start

Over 80% of enterprise AI initiatives fail before reaching production. The bottleneck is not model capability but context architecture. A technical analysis of the emerging discipline of context engineering.

Read
Mar 10, 2026
Security

Harvest Now, Decrypt Later: The Post-Quantum Cryptographic Threat to AI Infrastructure

Adversaries are stockpiling encrypted data today, betting that quantum computers will break current encryption within a decade. What this means for organizations running sensitive AI workloads.

Read
Mar 6, 2026
Infrastructure

Single Points of Failure: Concentration Risk in the Semiconductor Supply Chain

The global chip supply chain depends on a handful of irreplaceable facilities, materials, and monopolies. A technical assessment of concentration risk and its implications for AI compute sovereignty.

Read
Mar 2, 2026
Infrastructure

The Hardware Bottleneck: Physical Constraints on AI Scaling

AI progress is increasingly constrained not by algorithms but by the physical infrastructure required to run them. An analysis of power, cooling, chip lifespan, and supply chain fragility.

Read
Feb 26, 2026
Research

Agentic Systems and the Displacement of Enterprise Software

AI agents are not improving enterprise SaaS. They are replacing it. An analysis of the structural threat to per-seat licensing models and the emerging 'service-as-software' paradigm.

Read
Feb 20, 2026
Research

AI-Driven Labor Displacement and the Consumer Demand Feedback Loop

AI automation increases corporate efficiency but simultaneously reduces consumer purchasing power. An examination of the macroeconomic feedback loop that could constrain AI adoption.

Read
Feb 14, 2026
Research

Quantum Computing in 2026: Separating Engineering Progress from Cryptographic Threat

Google's Willow processor demonstrated a 13,000x speedup over classical computation. What this actually means for AI infrastructure, and what it does not mean for encryption.

Read