Blog
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 3517 posts
#2026-03 765#english 592#culture 264#deep-dive 254#kubernetes 247#career 229#ai 216#llm 208#devops 193#2026-04 146#security 141#database 114#observability 113#communication 109#history 107#architecture 100#productivity 96#finance 88#economy 84#mindset 81#psychology 80#ai-papers 79#food 78#it 78#travel 78#deep-learning 77#japanese 77#networking 77#performance 72#business-travel 70#linux 70#gpu 69#ai-agent 66#cs-fundamentals 63#postgresql 60#rag 58#self-improvement 55#learning 53#mlops 53#ai-platform 51
Open Source Worth Watching Right Now (2) Build Tools, Editors, CLIs, and Terminals
Work that gets repeated dozens of times a day, such as installing packages, linting, and bundling, has been rewritten in native languages, and the waiting has come down from seconds to milliseconds. This post introduces
2026-08-12 · 5 min read #open-source#developer-tools#cli#rust#performanceWhat Replaces Technology — And How to Check Whether Your Stack Sits on That Trajectory
The final part of this series on open source projects that gave up their place. Cutting across the thirty-odd projects covered in the previous three parts, it distils five forces that replace technology: platform absorpt
2026-08-12 · 7 min read #open-source#architecture#license#migration#governanceThe Generational Shift in Data Stores and Queues — Licenses, Forks and the Apache Attic
Ten projects in the data layer that gave up their place or had their distribution terms changed, documented using only official announcements and Apache Attic records as evidence. The license changes at Redis, Elasticsea
2026-08-12 · 9 min read #open-source#database#kafka#redis#elasticsearchGPU Serving SLOs and Alert Design — What to Promise and What to Page On
Putting an SLO on a GPU inference service starts with deciding which metric represents the user experience. Time to first token and throughput eat each other, so setting a target on only one guarantees the other collapse
2026-08-12 · 7 min read #gpu#kubernetes#slo#alerting#prometheusDevice Plugins and GPU Scheduling — Where nvidia.com/gpu Comes From
Kubernetes does not know what a GPU is. What makes a node advertise GPUs as a resource is a device plugin registered with the kubelet, and the name that falls out of it is the extended resource nvidia.com/gpu. This post
2026-08-12 · 6 min read #gpu#kubernetes#device-plugin#scheduling#nvidiaTranslating "It Is Slow" into an Engineering Problem — FDE Customer Communication
Customers do not file bug reports. They file pain reports: "it is slow," "it does not work," "sometimes it acts strange." One of the core skills of a Forward Deployed Engineer (FDE) is the questioning technique that tran
2026-08-12 · 5 min read #career#fde#forward-deployed-engineer#communication#customer-successFrom Backend, DevOps, or Data Engineering to FDE — a Six-Month Transition Roadmap
Most people weighing a move into Forward Deployed Engineer (FDE) work start from one of three places: backend, DevOps/SRE, or data engineering. The good news is that whichever it is, you already hold about half of the FD
2026-08-12 · 5 min read #career#fde#forward-deployed-engineer#career-transition#roadmapWhat Is a Forward Deployed Engineer (FDE) — the Engineer Stationed at the Customer Site
Palantir invented the title, and now OpenAI and Anthropic are racing to hire for it: Forward Deployed Engineer, or FDE. This post pins down what the role actually is — an engineer whose coordinate system is the customer
2026-08-12 · 6 min read #career#fde#forward-deployed-engineer#ai#job-searchWhy PoCs Never Reach Production — Success Criteria, Security Review, Handover
A large share of PoCs that earn applause at the demo die quietly anyway — not for lack of technology, but because there were no success criteria, the security review started last, no champion existed, or the gap between
2026-08-12 · 6 min read #career#fde#forward-deployed-engineer#poc#productionThe FDE Incident Diagnosis Playbook — Six Steps from Access to Report
The decisive difference between an incident on your own service and one at a customer site is that the latter starts from knowing nothing. That is why a Forward Deployed Engineer (FDE) needs a fixed order before needing
2026-08-12 · 6 min read #career#fde#forward-deployed-engineer#incident-response#debuggingThe FDE Skill Map — Baselines, Working Levels, and Check Questions for 8 Domains
A map of the skills a Forward Deployed Engineer (FDE) actually needs, drawn across eight domains: Linux, networking, Kubernetes, databases, auth and security, observability, cloud and infrastructure, and customer communi
2026-08-12 · 6 min read #career#fde#forward-deployed-engineer#skills#roadmapThe First 90 Days as an FDE — Mapping, Solo Tickets, and a Led Mission
Onboarding as a Forward Deployed Engineer (FDE) means adapting to two unfamiliar environments at once — your own company and the customer site — which is why it needs more deliberate design than a regular engineering onb
2026-08-12 · 5 min read #career#fde#forward-deployed-engineer#onboarding#checklistChoosing Speech Models: Practical Criteria for STT and TTS
Unlike text models, speech models are chosen after the language coverage, audio length constraints, real-time requirement, and diarization need are already fixed. This post lays out the parameters, licenses, language cov
2026-08-12 · 7 min read #ai#huggingface#open-source-llm#speech-to-text#text-to-speechChoosing Open Text Generation Models by Size Class
When picking an open text generation model, size is not a quality tier — it is a deployment constraint. This post splits the field into on-device small, single-GPU mid, and server-class large, listing the parameters, con
2026-08-12 · 7 min read #ai#llm#huggingface#open-source-llm#text-generationOpen Models That Support Korean, and the Cost of Tokenization
A line saying a model supports Korean means it processes the language, not that it is good at it. This post lays out the card values read on 2026-08-12 for Korean-focused, bilingual, and multilingual open models, explain
2026-08-12 · 6 min read #ai#llm#huggingface#open-source-llm#korean-nlpLicenses and Deployment: Reading the Types and Verifying Quantized Builds
The short identifier in an open model license field is a table of contents; the actual terms live in the full text. This post organizes the license types actually encountered on 2026-08-12, what community licenses and no
2026-08-12 · 7 min read #ai#llm#huggingface#open-source-llm#licenseChoosing Code Models: Completion vs Chat, FIM, and Licenses
Code models come in two different products: the completion model that fills the cursor position inside an editor, and the conversational model that answers questions. This post organizes the parameters, context, fill-in-
2026-08-12 · 7 min read #ai#llm#huggingface#open-source-llm#code-llmVision and Multimodal: Image Understanding, OCR, and VLMs
Vision models are not one category. Conversational VLMs, dedicated OCR, document structuring, and image retrieval embeddings are different objects and do not substitute for one another. This post organizes the card value
2026-08-12 · 7 min read #ai#huggingface#open-source-llm#vision-language-model#ocrHow to Read a Hugging Face Model Card: What to Trust and What to Doubt
Six fields on a Hugging Face model card actually carry decision weight: license, parameter count, context length, intended use, limitations, and training data. This post explains what each field means and what it hides,
2026-08-12 · 7 min read #ai#llm#huggingface#open-source-llm#model-cardEmbeddings and Rerankers: What Actually Matters in RAG
What decides your embedding model in a RAG build is not a leaderboard rank but dimensionality, maximum input length, prefix conventions, multilingual coverage, and the division of labor with a reranker. This post lays ou
2026-08-12 · 7 min read #ai#llm#huggingface#open-source-llm#embedding