Stack Builders - AI in Software Delivery: What’s Working, What’s Hard, and What We’re Still Learning

Contact Us

AI

Engineering Leadership

Project Management

Software Development

AI in Software Delivery: What’s Working, What’s Hard, and What We’re Still Learning

Share this blog

Copy Icon

Copy Check Icon

Stack Builders team

Jul. 1, 2026

Jul. 1, 2026

8 min read

Subscribe to blog

Email

Subscribe

Explore More Blog Posts

Cleaning Web Analytics: Identifying Bots with Gemini AI

Cleaning Web Analytics: Identifying Bots with Gemini AI

AI

Tutorial

Gisela Osorio

6 min read

AI Agents in Practice With OpenClaw

AI Agents in Practice With OpenClaw

AI

AI Agents

Stack Builders team

7 min read

When Text Becomes Code: Defending LLM–Database Integrations from Prompt Injection

When Text Becomes Code: Defending LLM–Database Integrations from Prompt Injection

Stack Builders team

10 min read

AI is changing software delivery, but not by removing the need for strong engineering judgment. This recap explores what Stack Builders’ senior tech leads are learning about AI-assisted workflows, code review, delivery metrics, estimation, and the process discipline needed to make AI useful in real projects.

AI-assisted development is no longer a side experiment. Across Stack Builders teams, senior technical leads are using AI in day-to-day delivery: generating components, supporting specification-driven workflows, reviewing larger pull requests, analyzing team metrics, and even coordinating parallel agents to resolve batches of issues.<br>But the most interesting insight from a recent Senior Tech Leads discussion was not simply “AI makes us faster.” The more useful takeaway was subtler: AI changes where the hard parts of software delivery live.<br>Writing code may be faster. Understanding the right request, measuring real progress, validating correctness, and keeping workflows aligned are becoming the new pressure points.<br>In this post, we’re sharing an exclusive recap of a strategic conversation with our senior tech leads, plus conversation questions your team can use to explore AI for software delivery with more confidence.<br>1. Specification-driven development is promising, but workflow alignment is still tricky<br>Several teams are experimenting with specification-driven development. One team compared a lighter-weight open-spec approach against a more structured framework that prescribes more of the workflow. The early signal was positive: specs help guide AI output and make implementation feel more controlled.<br>The catch? Synchronization.<br>When tasks live in a project tracker, tests live in a spec framework, and AI operates across both, teams start asking new questions:<br>Is the tracker still the source of truth?<br>Should tasks be declared closer to the repository?<br>How do we prevent specs, tickets, prompts, and PRs from drifting apart?<br>This is a familiar software quality problem wearing a new hat. AI does not remove the need for shared context. It makes stale context more expensive because the system can confidently accelerate in the wrong direction.<br>Conversation starter: Where should the source of truth live for AI-assisted work: the tracker, the repo, the spec, or some carefully stitched combination?<br>2. AI can reduce coding time while increasing cognitive load<br>One recurring theme was mental load. AI can generate more code, larger PRs, and broader solutions, but humans still need to understand what changed, why it changed, and whether it fits the domain.<br>One lead described using custom AI skills to explain large requests more clearly. Instead of only asking AI to produce code, the team used AI to research learning techniques and package them into a reusable skill that helps break down concepts, goals, non-goals, and tradeoffs.<br>That is a useful pattern: AI as a comprehension tool, not just a production tool.<br>The old bottleneck was often “Can we implement this?” The new bottleneck may be “Can we understand and validate this fast enough?”<br>Conversation starter: What workflows could help reviewers reduce cognitive load when reviewing AI-generated or AI-assisted code?<br>3. The best AI workflows may be project-specific, not generic<br>A front-end example made this clear. AI was helpful for generating design system components and CMS-backed sections from screenshots, but it was not perfect at interpreting Figma-style visuals. The team improved results by refining prompts and creating project-specific instructions.<br>Another lead described this as moving from “fix the AI’s output” to “fix the process.” Instead of repeatedly correcting the same mistakes manually, teams can update the harness: prompts, rules, memories, examples, restrictions, and validation loops.<br>That mindset is important. If the same AI mistake happens twice, it may not be a code problem. It may be a workflow design problem.<br>This aligns with Stack Builders’ broader AI positioning: AI should be applied where it drives value while preserving quality, security, and...