Want smarter AI? Don't rush to change models
Missing documents, forgotten requirements and unfinished work: a practical look at hybrid retrieval, context, inference budgets, agent verification and long-term memory.
# Artificial intelligence
17 articles
With DeepSeek's update, I want to know whether there will be less rework
DeepSeek Search and Fetch fit into a writing workflow. What matters to me is reading the sources, checking the facts and saving time all the way through revision.
AI is splitting up the work. I'm more interested in who finishes the job
ChatGPT's parallel agents, Cursor's cloud environments, Gemini's Minecraft team and Claude's quantum circuits: what I'd want before using them in my own projects.
The More I Develop with AI, the Less I Skip the Process
What improving my own blog taught me about AI development: clear requirements, small changes, meaningful tests and live verification.
AI Briefing, October 15: Grok Voice 3.0 and the Full Cost of Customer Support
Grok Voice 3.0 claims 70% autonomous resolution of Starlink support calls. Resolution, handoffs and costs, plus OpenAI adoption data and Argon research.
Is Gemini 4 Argon Worth Using? Complex Code, Long Tasks and Real Costs
Where Gemini 4 Argon fits: engineering, documents and agents, with access requirements, shared context limits, introductory pricing and a practical trial.
AI Briefing, October 14: Gemini 4 Argon and Changing Agent Workflows
Gemini 4 Argon, Anthropic’s source control, LangSmith AI Search and three quick updates, with practical terminology and source posts.
WorkBuddy or Codex? Choose by the Work You Need Done
A practical guide to office files, code projects, better task instructions, handoffs and the real cost of using WorkBuddy and Codex.
Making a 30-Second Ad: Which AI Video Model Should You Choose?
Pick a model for product photos, reference-driven shots, real-footage edits or audio. Includes a 30-second shot plan and a practical API budget example.
Which Claude Code Model Should You Use? Let’s Talk Tasks and Costs
Choose between Sonnet 5.5, Opus 5.5, Haiku 5.5 and Fable 5.1 by the work you need done, with current API prices, context tiers, effort and caching costs.
Langfuse in Practice: Make Slow, Expensive and Wrong AI Responses Diagnosable
Instrument a small RAG workflow with Python SDK v4, connect traces to prompt versions and evaluations, and understand self-hosting, masking and real task costs.
Gemini Spark: An Agent That Keeps Working After the Answer
What Gemini Spark’s Tasks, Skills and Schedules do, how background execution works, who can access it, and which everyday workflows are worth trying first.
Claude’s New Report: Capable AI Needs to Know When to Stop
Anthropic reports unintended actions during Claude evaluations and internal use. Lessons for agent permissions, stopping rules and reliable automation.
Is GPT-6.1 Sol Worth Switching To? Speed, Uses and Savings over GPT-5.6
An October 9, 2026 comparison of GPT-6.1 Sol, GPT-5.6 Sol, Astra and Luna, explaining speed tiers, reasoning effort, API prices and practical task costs.
Choosing an AI Video Model: Get the Shot Right Before Chasing a Cinematic Look
Choose between Seedance, Runway, Google and Wan by the shot you need. Reduce retries and keep real footage for reporting on vehicles and driving features.
GPT, Claude or Gemini? Comparing Five AI Model Families and Their Costs
A dated comparison of GPT, Claude, Gemini, DeepSeek and Qwen: inputs, tools, API costs and limits, with practical criteria for writing, coding and analysis.
FDE, Part 1: What a Forward Deployed Engineer Does in Enterprise AI
How FDEs connect business needs, enterprise data and software capabilities, with clear responsibilities and delivery boundaries.