Unix Philosophy, Named by the Document You Quote
The Unix philosophy is four 1978 rules for small tools, not one slogan. Name the document you quote, then use those rules when you build with agents and tools.
Read
Author
Articles by Tomas Laurinavicius
Tomas builds AI-native software and media products at Craftled, including Bordfeed, Epigraph Media and Laupix, a fully autonomous zero-human company. He writes about AI from the builder's seat, not the sidelines.
More Articles
RAG: Retrieval-Augmented Generation, Pipelines, Skip Rules
Retrieval-augmented generation (RAG) pairs a frozen language model's parametric memory with a retrievable non-parametric corpus.
Read
NIST AI Risk Management Framework as a Working Operating System
Voluntary NIST AI 100-1 guidance (26 January 2023). Map GOVERN, MAP, MEASURE, and MANAGE to owners, evidence, and cadence.
Read
ChatGPT Enterprise Starts With the Control Plane
Buyer and implementation guide for ChatGPT Enterprise: controls, SSO/SCIM, residency, HIPAA boundary, and a 12-step rollout.
Read
Why Run a Local LLM If ChatGPT Already Works?
Run a local LLM on hardware you control. Covers privacy vs offline, VRAM tiers, Q4_K_M, Ollama vs vLLM, and TCO as utilization rather than one crossover.
Read
Why Does Enterprise AI Stall? A Portfolio Playbook
Run enterprise AI as a Fund, Defer, or Kill portfolio. Covers operating models, platform layers, shadow AI, and conflicting ROI numbers.
Read
AI Governance Built on Owners, Inventory, Cadence
AI governance is named owners, written policy, risk-tiered approvals, a living inventory, and a review cadence that continues after go-live.
Read
Prompt Engineering: Specs, Evals, and Team Workflow
Prompt engineering is how you spec, evaluate, and version LLM behavior in production. Covers CoT caveats, prompt injection, and when RAG or fine-tuning wins.
Read
OpenAI API in Production: Keys, Responses, Rate Limits
The OpenAI API is a billed HTTP developer product. This guide covers project keys, Responses vs Completions, model pinning, and 429s.
Read
LLM Observability: Instrument, Score, Contain
LLM observability records traces, quality scores, cost, and latency so you can explain a bad output and contain silent failures.
Read
Claude vs ChatGPT: Which One Actually Fits How Your Team Works?
Claude wins on coding, long documents, and privacy defaults; ChatGPT wins on multimodal breadth, browsing, and integrations. Full breakdown across 7 workflows.
Read
12 Tested AI Voice Assistants for 2026: Personal Use and Business Deployment
Compare 12 AI voice assistants for personal and business use. Real pricing, latency benchmarks, and true cost breakdowns for Siri, Vapi, Retell AI, and more.
Read
Showing 1 - 12 of 33 articles