0>1.software

// blog

Blog

Insights on AI, software development, and the future of work.

How Many Tokens Does Claude Code Send Per Request?
John Sasser

Proxy-level measurements of Claude Code token overhead: the 33k request floor, cache write behavior, instruction file cost, and subagent fan-out.

John Sasser
July 13, 2026
6 min read
claude-codeopencodeagentic-codingtoken-costsprompt-caching
How Claude Code's Hidden Prompt Markers Work
John Sasser

Claude Code 2.1.196 embeds steganographic markers in its system prompt based on ANTHROPIC_BASE_URL and timezone. How the mechanism works and who it flags.

John Sasser
July 6, 2026
6 min read
claude-codesecuritysteganographyai-toolingprivacy
GPT-5.6 Sol Preview: Models, Benchmarks, Gated Launch
John Sasser

What OpenAI's GPT-5.6 Sol preview includes: the Sol, Terra, and Luna lineup, benchmark claims, cyber safeguards, and why the launch starts gated.

John Sasser
June 29, 2026
6 min read
openaigpt-5.6ai-benchmarksai-policyagentic-coding
Fable 5 Export Control: What Actually Happened
John Sasser

How a 'fix this code' prompt led the US to export control Anthropic's Fable 5 and Mythos 5, and why security researchers say the ban hurts defenders.

John Sasser
June 22, 2026
6 min read
export-controlsanthropicai-securityappsecpolicy
Why the US Government Suspended Fable 5 and Mythos 5
John Sasser

How a US export control directive forced Anthropic to suspend Fable 5 and Mythos 5, what the disputed jailbreak actually was, and why the process matters.

John Sasser
June 15, 2026
6 min read
anthropicexport-controlsai-policyjailbreaksfable-5
Which Software Engineering Skills Still Matter With AI
John Sasser

A payments engineer with 10 years of experience says AI eroded his expertise. A close look at which software engineering skills still matter in 2026.

John Sasser
June 8, 2026
5 min read
aisoftware-engineeringcareersagentic-codingllms
Claude Opus 4.8: What Changed for Agent Builders
John Sasser

What changed in Claude Opus 4.8: effort controls, dynamic workflows in Claude Code, mid-task system prompts in the Messages API, and unchanged pricing.

John Sasser
June 1, 2026
7 min read
claudeanthropicclaude-codeagentic-aillm
DeepSeek Prefix Caching for Coding Agents
John Sasser

DeepSeek prefix caching can lower coding-agent input cost. Learn how Reasonix preserves cache hits, where it helps, and its operating limits.

John Sasser
May 25, 2026
8 min read
deepseekcoding-agentsprefix-cachingagentic-ai
Python or Rust for AI-Assisted Coding
John Sasser

Python or Rust for AI-assisted coding depends on feedback loops, runtime constraints, ecosystem fit, and the agent's language quality.

John Sasser
May 18, 2026
8 min read
pythonrustai-codingsoftware-engineering
Local AI vs Cloud AI for App Features
John Sasser

Local AI vs cloud AI for app features: choose on-device models for private data transformation, typed outputs, and fewer service dependencies.

John Sasser
May 11, 2026
8 min read
local-aicloud-aiappleswiftai-engineering
Why Claude Code Refuses Repos That Mention OpenClaw
John Sasser

How Claude Code ingests git history into the model's context, why a commit mentioning OpenClaw can trigger refusals or extra charges, and how to debug it.

John Sasser
May 4, 2026
5 min read
claude-codeagentic-aicontext-engineeringai-safetydeveloper-tools
Claude Code Quality Issues: What Happened
John Sasser

Claude Code quality issues in April 2026 came from reasoning defaults, lost session context, and a prompt constraint. Learn the fixes.

John Sasser
April 27, 2026
8 min read
claude-codeai-engineeringagentic-aiprompting
Claude Opus 4.7 for Long-Running Coding Tasks
John Sasser

Claude Opus 4.7 improves long-running coding tasks with stronger validation, tool use, vision, and instruction following at Opus 4.6 pricing.

John Sasser
April 20, 2026
8 min read
claudeai-engineeringcoding-agentssoftware-development
How to Detect Claude Code Quality Regressions in Your Logs
John Sasser

What a data-mined GitHub issue reveals about measuring Claude Code quality regressions from session logs, and why extended thinking depth matters.

John Sasser
April 13, 2026
6 min read
claude-codeagentic-aillm-observabilityextended-thinkingai-engineering
How Source Maps Leak Source Code From npm Packages
John Sasser

Source maps published to npm can expose complete unminified source code. How the Claude Code leak happened and how to audit your packages for .map files.

John Sasser
April 6, 2026
5 min read
source-mapsnpmclaude-codejavascriptsupply-chain
Python .pth File Attacks and the litellm Compromise
John Sasser

How the malicious litellm_init.pth in litellm 1.82.8 stole credentials on Python startup, how .pth code execution works, and how to check and remediate.

John Sasser
March 30, 2026
6 min read
supply-chainpythonsecuritylitellmcredential-theft
Leanstral: Mistral's Open-Source Lean 4 Proof Agent
John Sasser

What Leanstral is, how Mistral's open-source Lean 4 proof agent works, its FLTEval scores against Claude, and why Lean makes pass@k sampling pay off.

John Sasser
March 23, 2026
6 min read
leanstrallean4formal-verificationmistralagentic-ai
Claude 1M Context Window Pricing and Limits Explained
John Sasser

Claude Opus 4.6 and Sonnet 4.6 now offer the full 1M context window at standard pricing. What it costs, how recall holds up, and when to use it.

John Sasser
March 16, 2026
6 min read
claudelong-contextllm-pricingagentscontext-engineering
Why Veteran Developers Split on AI Coding Agents
John Sasser

Why experienced developers are split on AI coding agents like Claude Code, and what the old fight over CUDA adoption predicts about how it resolves.

John Sasser
March 9, 2026
6 min read
ai-coding-agentsclaude-codedeveloper-experienceagentic-ai
OpenAI Classified Network Deal: What Was Agreed
John Sasser

What OpenAI's Department of War agreement covers: classified network deployment, cloud-only hosting, forward-deployed engineers, safety terms.

John Sasser
March 2, 2026
7 min read
openaidefense-aiclassified-networksai-policygovtech
Claude Sonnet 4.6 Migration Guide
John Sasser

Claude Sonnet 4.6 migration guidance for choosing thinking effort, testing long-context work, and safely upgrading coding and agent workflows.

John Sasser
February 23, 2026
9 min read
claudesonnetai-engineeringagentic-aillms
Why AI Agents Violate Constraints Under KPI Pressure
John Sasser

A new benchmark finds AI agents violate ethical constraints in up to 62.8% of KPI-driven scenarios, with safety not improving across model generations.

John Sasser
February 16, 2026
5 min read
ai-agentsalignmentbenchmarksagent-safetyllm-evaluation
How Parallel Claude Agents Built a C Compiler
John Sasser

How Anthropic ran 16 parallel Claude agents to build a 100,000-line C compiler: the harness design, test strategy, and where agent teams hit limits.

John Sasser
February 9, 2026
6 min read
agentsclaudeharness-designcompilersautonomous-coding
AGENTS.md vs Skills for Teaching Agents Framework Docs
John Sasser

Vercel's evals found an 8KB docs index in AGENTS.md scored 100% on Next.js 16 tasks while skills peaked at 79%. Here is why passive context won.

John Sasser
February 2, 2026
6 min read
agents-mdskillscoding-agentscontext-engineeringnextjs
Claude Constitution: How It Works
John Sasser

Claude Constitution explains how Anthropic uses a public values document in training, where its priorities apply, and what it cannot guarantee.

John Sasser
January 26, 2026
7 min read
claudeanthropicconstitutional-aiai-safetymodel-training
Will AI Replace Software Developers? A 50-Year Pattern
John Sasser

Will AI replace software developers? Fifty years of attempts, from COBOL to no-code, show why the bottleneck is judgment and what actually changes.

John Sasser
January 19, 2026
5 min read
ai-codingsoftware-engineeringdeveloper-productivitylow-code
How Coding Agents Work
John Sasser

How coding agents work: the tool loop, file operations, prompts, and safety boundaries behind a small but functional coding assistant.

John Sasser
January 12, 2026
8 min read
coding-agentsclaude-codellm-toolspython
How Much Claude Code Customization Do You Need
John Sasser

Boris Cherny, creator of Claude Code, runs a near-vanilla setup. When Claude Code customization like CLAUDE.md, hooks, and MCP servers actually pays off.

John Sasser
January 5, 2026
5 min read
claude-codeai-codingagentic-aideveloper-toolscontext-engineering
AI in Recruiting: The Gap Between Perception and Reality

What both job seekers and talent professionals get wrong about AI-powered resume screening

John Sasser

Job seekers are convinced AI is automatically rejecting their applications, sometimes illegally. Talent professionals insist AI screening features barely exist, and they personally review every candidate. Both are partially right. Both are partially wrong.

John Sasser
January 5, 2026
10 min read
AIRecruitingHR Tech