Free playbook: A practical operating model for AI across the SDLC

Planning software for AI coding agents

Product thinking, technical deep-dives, and lessons on planning software for agentic development.

17 articles tagged “Spec Governance”

Audit trail diagram for AI-generated code: ticket, approved spec, agent handoff, and commit SHA linked as evidence
Spec Governance12 min read

Audit Trails for AI-Generated Code in Regulated Teams

When an auditor asks who approved an AI-built feature, git history cannot answer. Here is what an audit trail for AI-generated code must capture, and how to produce it at decision time.

Dishant Sethi

Dishant Sethi

November 2, 2026
Diagram comparing a PRD that captures product intent with an agent-ready spec of scope, acceptance criteria, approval gate
Product Management14 min read

Spec vs PRD in the Age of AI Coding Agents

A PRD captures product intent for people; a spec is the execution contract AI coding agents build from. Here is how they differ, and why high-functioning teams keep both.

Dishant Sethi

Dishant Sethi

October 30, 2026
Diagram of the three AI-DLC phases Inception, Construction, and Operations with human approval gates between each phase
AI Development11 min read

AI-DLC Phases: Inception, Construction, and Where Humans Gate

The AI-DLC phases (Inception, Construction, Operations) reorder delivery around agents that draft and humans who approve. Here is what each phase produces, and where the human gates belong.

Dishant Sethi

Dishant Sethi

October 28, 2026
Architecture diagram showing AI coding agents across repositories compared to unified cross-service feature planning
AI Development9 min read

AI Coding Agents in Multi-Repo Codebases: Where Plans Break

AI coding agents are built for single repositories, but enterprise features span several. Here is how to coordinate agents across multi-repo microservices.

Dishant Sethi

Dishant Sethi

October 26, 2026
Comparison spectrum of Spec Kit vs Kiro vs Tessl showing CLI scaffolding, IDE specs, and spec as source architectures
AI Development11 min read

Spec Kit vs Kiro vs Tessl: What Each Assumes About Your Team

Three spec-driven development tools compared through the team assumptions each makes: authoring, review responsibilities, and multi-repo governance.

Dishant Sethi

Dishant Sethi

October 23, 2026
Architecture diagram showing how to prevent spec drift across autonomous AI coding agents using a unified plan
AI Development9 min read

How to Prevent Spec Drift When Multiple Agents Share a Codebase

When multiple AI agents edit the same codebase, specifications and source code diverge. Here is how to prevent spec drift with unified planning.

Dishant Sethi

Dishant Sethi

October 21, 2026
Diagram showing a PM reviewing the agent plan (problem, scope, criteria) instead of a 600-line code diff
Product Management11 min read

Reviewing AI-Generated Code as a PM: Review the Plan, Not the Diff

When devs use coding agents, PMs are often asked to review the finished diff. Here is how to review AI-generated code by gating the plan instead of the syntax.

Dishant Sethi

Dishant Sethi

October 19, 2026
Architecture diagram showing one shared AGENTS.md spec file driving Cursor, Claude Code, and GitHub Copilot without drift
AI Development9 min read

Spec-Driven Development with Cursor, Claude Code, and Copilot

Cursor, Claude Code, and Copilot each read distinct configuration files. Here is how to configure a unified spec layer across all three coding assistants.

Dishant Sethi

Dishant Sethi

October 16, 2026
Workflow diagram of an AI spec approval gate where product, engineering, and security sign off before coding
Spec Governance11 min read

Spec Approval Workflows for AI Coding Agents: Who Signs Off

A spec approval workflow decides who reviews an AI coding agent's plan, who signs off before it builds, and how approvals are recorded for compliance.

Dishant Sethi

Dishant Sethi

October 14, 2026
← Prev12Next →