Medium

Josphine Job

Pinned

In

Artificial Intelligence in Plain English

by

Josphine Job

·

Jul 30

AI Agent Guardrails for Prompt Injection Defense

How a sentinel file connects a detection hook to an enforcement hook, and what it takes to stop injected instructions from becoming…

AI Agent Guardrails for Prompt Injection Defense

Pinned

In

Artificial Intelligence in Plain English

by

Josphine Job

·

Jul 21

How I Built Persistent AI Agent Memory Without a Database

A simple method to keep agents context-aware, cost-efficient, and maintainable across long projects.

How I Built Persistent AI Agent Memory Without a Database

Pinned

Josphine Job

·

Jun 9

How to Build a Self-Managing Multi-Agent AI Workflow

You provide the requirement, and a team of AI agents handles the rest without you orchestrating each step.

How to Build a Self-Managing Multi-Agent AI Workflow

Pinned

In

Artificial Intelligence in Plain English

by

Josphine Job

·

Jul 10

How to Measure What Your AI Session Actually Costs

Building a Token Cost Analyzer to Audit Your AI Coding Workspace Before It Costs You

How to Measure What Your AI Session Actually Costs

Pinned

In

Artificial Intelligence in Plain English

by

Josphine Job

·

Apr 11

Why Every QA Should Build Their Own AI Workspace

Define your QA standards once, apply them with any AI tool

Why Every QA Should Build Their Own AI Workspace

In

Level Up Coding

by

Josphine Job

·

6d ago

How to Keep AI-Generated PRs Small Enough for Humans to Review

The strategies I tried to keep pull requests within the review limit during spec-driven development.

How to Keep AI-Generated PRs Small Enough for Humans to Review

In

Level Up Coding

by

Josphine Job

·

Nov 6

What a Good Spec Looks Like in Spec-Driven Development

The format I use, why each section exists, and what I lost on a project where I skipped it.

What a Good Spec Looks Like in Spec-Driven Development

In

Women in Technology

by

Josphine Job

·

Nov 6

Why an Agentic Test Pipeline Can't Replace a Human QA

What surprised me about agentic testing, and the things I learnt along the way.

Why an Agentic Test Pipeline Can't Replace a Human QA

In

Level Up Coding

by

Josphine Job

·

Nov 4

How I Reduced the Time and Cost of AI Agent Code Search on a Large Codebase

Why a semantic code graph index cut token cost and search time on a large brownfield rebuild, and how I estimated the savings.

How I Reduced the Time and Cost of AI Agent Code Search on a Large Codebase

In

Artificial Intelligence in Plain English

by

Josphine Job

·

Nov 3

Why My Agentic Test Pipeline Is Not Exploratory Testing

I could have built an agent that explores like a human tester, but I chose determinism and lower token cost instead.

Why My Agentic Test Pipeline Is Not Exploratory Testing

Josphine Job

Josphine Job

Software engineer writing about AI native development, test automation, and software testing. Follow if you learn better from hands-on stories than text books.

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech