agents
- Feedback That Changes Behaviour Is Feedback You Can Check — July 1, 2026
ai
- When Nobody Reads the Code Anymore — March 15, 2026
- One Agent Per Domain, Zero Trust — March 8, 2026
ai-assisted-development
- Feedback That Changes Behaviour Is Feedback You Can Check — July 1, 2026
- Hard gates don't break whoever holds them. They break the weaker model. — June 24, 2026
- Don't let a model grade its own family — June 20, 2026
- Spec Kit's fit and limits in ERP modernization — May 22, 2026
- Implementing the spec: where discipline holds and where it slips — May 12, 2026
- Auditing as you go: what corrections actually look like — May 5, 2026
- Spec Kit on a brownfield codebase: setup and first impressions — May 3, 2026
ai-tooling
- Why the Smartest AI Agent Is the Worst Translator — February 17, 2026
brownfield
- Hard gates don't break whoever holds them. They break the weaker model. — June 24, 2026
- Don't let a model grade its own family — June 20, 2026
- Spec Kit's fit and limits in ERP modernization — May 22, 2026
- Implementing the spec: where discipline holds and where it slips — May 12, 2026
- Auditing as you go: what corrections actually look like — May 5, 2026
- Spec Kit on a brownfield codebase: setup and first impressions — May 3, 2026
c
- Quantifying a Legacy Codebase You Can't Rewrite — April 15, 2026
- When You Can't Embed, Bridge — March 7, 2026
- When "It Compiled" Is the Dangerous Part — March 6, 2026
- Y2038: When "Impossible" Means "Wrong Approach" — March 5, 2026
code-review
- When Nobody Reads the Code Anymore — March 15, 2026
- Local LLMs for Code Review: Why I Don't Send Proprietary Code to the Cloud — February 28, 2026
database
- Quantifying a Legacy Codebase You Can't Rewrite — April 15, 2026
engineering-culture
- Grade Your Evidence, Not Your Confidence — July 3, 2026
- Feedback That Changes Behaviour Is Feedback You Can Check — July 1, 2026
erp-modernization
- Spec Kit's fit and limits in ERP modernization — May 22, 2026
evaluation
- Hard gates don't break whoever holds them. They break the weaker model. — June 24, 2026
- Don't let a model grade its own family — June 20, 2026
- Spec Kit's fit and limits in ERP modernization — May 22, 2026
- Implementing the spec: where discipline holds and where it slips — May 12, 2026
- Auditing as you go: what corrections actually look like — May 5, 2026
- Spec Kit on a brownfield codebase: setup and first impressions — May 3, 2026
feedback
- Feedback That Changes Behaviour Is Feedback You Can Check — July 1, 2026
ipc
- When You Can't Embed, Bridge — March 7, 2026
knowledge-management
- When Nobody Reads the Code Anymore — March 15, 2026
legacy-migration
- One Agent Per Domain, Zero Trust — March 8, 2026
- When You Can't Embed, Bridge — March 7, 2026
- Parity Testing: The Only Definition of Correct — February 24, 2026
- Teaching an LLM a Language It Has Never Seen — February 20, 2026
- Why the Smartest AI Agent Is the Worst Translator — February 17, 2026
legacy-modernization
- Quantifying a Legacy Codebase You Can't Rewrite — April 15, 2026
legacy-systems
- Conservative by Design: Transpiling Thousands of Legacy Scripts Without Guessing — March 25, 2026
- When Nobody Reads the Code Anymore — March 15, 2026
- When You Can't Embed, Bridge — March 7, 2026
- When "It Compiled" Is the Dangerous Part — March 6, 2026
- Y2038: When "Impossible" Means "Wrong Approach" — March 5, 2026
llm
- Local LLMs for Code Review: Why I Don't Send Proprietary Code to the Cloud — February 28, 2026
- Teaching an LLM a Language It Has Never Seen — February 20, 2026
- Why the Smartest AI Agent Is the Worst Translator — February 17, 2026
llm-as-judge
- Don't let a model grade its own family — June 20, 2026
llm-reliability
local-inference
- Local LLMs for Code Review: Why I Don't Send Proprietary Code to the Cloud — February 28, 2026
lp64
- When "It Compiled" Is the Dangerous Part — March 6, 2026
methodology
- Grade Your Evidence, Not Your Confidence — July 3, 2026
migration
- Quantifying a Legacy Codebase You Can't Rewrite — April 15, 2026
- Conservative by Design: Transpiling Thousands of Legacy Scripts Without Guessing — March 25, 2026
multi-agent
- One Agent Per Domain, Zero Trust — March 8, 2026
prompt-engineering
- Teaching an LLM a Language It Has Never Seen — February 20, 2026
python
- Conservative by Design: Transpiling Thousands of Legacy Scripts Without Guessing — March 25, 2026
- When You Can't Embed, Bridge — March 7, 2026
rag
- Local LLMs for Code Review: Why I Don't Send Proprietary Code to the Cloud — February 28, 2026
spec-kit
- Hard gates don't break whoever holds them. They break the weaker model. — June 24, 2026
- Don't let a model grade its own family — June 20, 2026
- Spec Kit's fit and limits in ERP modernization — May 22, 2026
- Implementing the spec: where discipline holds and where it slips — May 12, 2026
- Auditing as you go: what corrections actually look like — May 5, 2026
- Spec Kit on a brownfield codebase: setup and first impressions — May 3, 2026
strangler-fig
- Quantifying a Legacy Codebase You Can't Rewrite — April 15, 2026
systems-programming
- When "It Compiled" Is the Dangerous Part — March 6, 2026
- Y2038: When "Impossible" Means "Wrong Approach" — March 5, 2026
testing
- Parity Testing: The Only Definition of Correct — February 24, 2026
transpiler
verification
- Grade Your Evidence, Not Your Confidence — July 3, 2026
- Parity Testing: The Only Definition of Correct — February 24, 2026
y2038
- Y2038: When "Impossible" Means "Wrong Approach" — March 5, 2026