Blog

Posts on AI workflow engineering: software that executes work, with deterministic logic, AI agents and human-in-the-loop in one system. Plus agentic development as well as app and web development with React and React Native.

AI Agentic Development

How AI coding agents become a reliable part of the engineering workflow: clear rules, tests, evals and reviews instead of luck. Practice from the terminal to the pipeline.

AI Agentic Development14 November 2026
Engineering the Harness: A Practical Architectural Pattern for AI Coding Agents in Production
How production engineering harnesses solve local model drift and system-wide architectural failures when deploying AI coding agents.
AI Agentic Development13 November 2026
Scaling AI Code Review: Multi-Agent Architecture and Production Acceptance Trade-Offs
How engineering teams build production AI code review agents using a multi-agent fan-out architecture to balance precision and engineer acceptance.
AI Agentic Development10 November 2026
Agentic Workflows for Scheduled Code Audits: Implementation Architecture and Token Costs
Scheduled agentic workflows can run daily read-only audits for code structure violations while exposing token usage metrics. This operational pattern surfaces structural drift before it compounds into technical debt.
AI Agentic Development05 November 2026
VS Code 1.145 Expands Remote Agent Workflows and Nested Coding Agent Harness Controls
Visual Studio Code 1.145 introduces remote browser support, nested coding agent harnesses, and local cron-driven workspace automations. These updates let engineering teams validate agent changes inside integrated dev previews without manual browser switching.
AI Agentic Development05 November 2026
Scaling AI Coding Agents in Production Codebases Without Losing Quality
Deploying coding agents to production requires strict task sizing, onboarding instruction files, and test suites to prevent hidden security gaps.
AI Agentic Development31 October 2026
AI Code Review Bottlenecks: Managing Pull Request Volume Growth Under Agentic Coding
Agentic coding accelerates pull request volume beyond human reviewer capacity. Engineering teams must implement automated risk-based triage to handle code review bottlenecks.
AI Agentic Development30 October 2026
Architecting Production AI Coding Agents for Automated Pull Request Review
Build reliable multi-agent systems for code review using parallel comment generators, vector search, and strict filtering pipelines.
AI Agentic Development29 October 2026
AI Coding Agents in Production: Solving the Code Review Bottleneck
Recent industry data reveals that while AI coding agents handle massive shares of daily development tasks, overall software output fails to scale at the same rate. Engineering teams must shift focus from code generation speed to strict review pipelines and quality gates.
AI Agentic Development29 October 2026
VS Code 1.144 Implements Copilot Fusion Multi-Model Orchestration for AI Code Review Workflows
Visual Studio Code 1.144 introduces Copilot Fusion multi-model orchestration for automated code review tasks. Teams must navigate distinct effort tiers and per-model billing models across environments.
AI Agentic Development26 October 2026
Why AI Coding Agents Increase Code Volume Without Improving Software Development Efficiency
An analysis of 300 million work events shows that AI agents generate more code and commits, but create severe downstream review bottlenecks that stall overall delivery.
AI Agentic Development24 October 2026
AI Coding Agents in Production: Enforcing Schema Validation and Hard Infrastructural Guardrails
Moving AI coding agents from demos to production requires strict schema validation, credential boundaries, and hard spending caps rather than polite prompt instructions.
AI Agentic Development22 October 2026
Mitigating Secret Leaks and Permission Failures in Production AI Coding Agents
Running AI coding agents against live fintech infrastructure reveals recurring operational failure modes. Mitigate them using tight task scopes, explicit allow policies, and human gates for deploys.
AI Agentic Development21 October 2026
QAOps and CI/CD: How Quality Gates Prevent Production Failures in Software Engineering
How integrating quality gates into CI/CD pipelines moves quality assurance from a final release checkpoint to a continuous engineering strategy.
AI Agentic Development18 October 2026
State of AI Code Review: Why Only 33 Percent of Suggestions Become Code
Production telemetry across 180,739 AI code review suggestions reveals that only 33.2 percent land as code. Engineering teams must build stricter quality gates to filter noisy findings.
AI Agentic Development18 October 2026
AI Coding Agents in Production: Scoped Delegation and Executable Test Verification
Modern AI coding agents require precise repository boundaries and strict test verification instead of sprawling prompts. Tech leads can maintain engineering control through scoped task delegation.
AI Agentic Development17 October 2026
HumanLayer CRISPY Framework for AI Coding Agents in Production
HumanLayer replaces monolithic RPI prompts with the seven-stage CRISPY framework to improve reliability and code ownership when running AI coding agents in production.
AI Agentic Development15 October 2026
Mitigating Business Logic Vulnerabilities and Supply Chain Risks in AI Coding Agents
AI coding agents generate code faster than human teams can review, introducing severe access control flaws and malicious open-source packages. Architecture teams need automated pre-review gates to catch logic risks.
AI Agentic Development14 October 2026
Software Development Efficiency and the 5.25x Review Bottleneck in AI Pull Requests
LinearB's 2026 data shows AI-generated pull requests wait 17.6 hours for review compared to 3.4 hours for unassisted code. Discover how decomposition pinpoints pipeline bottlenecks.
AI Agentic Development12 October 2026
Predictive Quality Gates in CI/CD Pipelines: Preventing Technical Debt and Build Bottlenecks
How to embed predictive quality gates into your CI/CD pipeline using CodeScene integrations. Discover how automated code health analysis prevents technical debt without slowing down pull requests.
AI Agentic Development10 October 2026
Engineering a Security Harness for AI Coding Agents
Deploying AI coding agents without a security harness introduces critical risks to source code and infrastructure. Discover how to build effective guardrails.
AI Agentic Development08 October 2026
Reviewing Code from AI Coding Agents in Production Requires Granular Risk and Confidence Signals
AI coding agents generate code faster than engineers can review it. A new research framework proposes exposing risk and confidence signals at the exact granularity of developer attention.
AI Agentic Development06 October 2026
Automated E2E Testing Bottlenecks: Framework Adoption Realities and Maintenance Trade-Offs
Analyzing industry data on automated E2E testing adoption, framework shifts toward Playwright, and why most engineering teams still lack critical path coverage.

PROJECT INQUIRY

Automate a business process?

Tell us which process currently takes too much time or manual effort. Together we examine what an executable workflow for it looks like, and reply personally.

info@appamass.com