Papers Added Oct 15, 2026
What shapes developers’ trust in AI coding tools?
A survey of 238 developers links trust in generative AI to output quality, practical value, and staying aligned with task goals.
Read more →Research
Go deeper when you're ready 📚
Reports, honest tool comparisons, and a jargon-busting glossary. For the concepts, infrastructure and communities behind it all, see The Vibe Coding Toolkit.
Papers Added Oct 15, 2026
AI Helped a Student Team Reduce Agile Workflow Overhead
A six-month software project found AI useful for task planning, technology research, and onboarding, while shared review helped preserve developer ownership.
Read more →Papers Added Oct 15, 2026
One-developer case study finds faster code output with an AI agent
An experienced developer produced code 2.34 times faster with one AI agent, but the case study does not establish equivalent gains in quality or overall productivity.
Read more →Papers Added Oct 15, 2026
One Developer Built Five App Ports with AI and a Shared Specification
A case study explores AI-assisted development using an executable specification and parallel implementations to check behavior across five app ports.
Read more →Papers Added Oct 15, 2026
A model-driven framework for more reliable LLM agent systems
MDE-Agentware turns structured agent designs into executable systems, with reported improvements in coordination, resilience, fairness, and energy use.
Read more →Papers Added Oct 15, 2026
AI code research pays less attention to efficiency and maintainability
A review finds that AI code evaluation focuses on correctness and security, while efficiency and maintainability lack comparable benchmarks and improvement methods.
Read more →Papers Added Oct 15, 2026
AI security fixes need verification, not just successful execution
A review of 53 studies supports bounded, auditable AI use in vulnerability management—not unrestricted autonomous security work.
Read more →Papers Added Oct 15, 2026
Connecting an AI assistant to a database without extra model training
The authors tested a database-backed AI agent that uses structured text, two-stage prompting, and output checks to answer pharmacology questions.
Read more →Papers Added Oct 15, 2026
How MCP Connects AI Coding Agents to Databases
The authors explore how agents use MCP to interact with databases, including a simple PostgreSQL server for database management and SQL execution.
Read more →Papers Added Oct 15, 2026
Understanding LLMs: Training, Prompting, Agents, and Evaluation
This survey maps how language models are trained, adapted, used, and evaluated, offering background for builders working with AI tools.
Read more →Papers Added Oct 15, 2026
Using Feedback Loops to Check AI-Generated Security Fixes
The authors study an LLM agent that repeatedly detects, fixes, and checks code vulnerabilities, using verification feedback to guide the next pass.
Read more →Papers Added Oct 15, 2026
Designing LLMs as collaborators, not standalone controllers
A review of automated-driving research suggests pairing LLMs with verified specialist modules rather than letting them control safety-critical systems alone.
Read more →Papers Added Oct 15, 2026
Generating web automation code from a successful browser session
ASG-LLM combines browser traffic, selective function retrieval, and execution feedback to generate scraping code for security-sensitive web services.
Read more →Papers Added Oct 15, 2026
Better specifications mattered more than adding a second coding agent
A controlled study of LLM code generation found that specification quality mattered more than single-agent versus two-agent architecture.
Read more →Papers Added Oct 15, 2026
What a Public Dataset Can Tell You About AI Coding Workflows
A release supports reanalysis of a controlled AI coding workflow study, but missing code and tests limit what builders can independently verify.
Read more →Papers Added Oct 15, 2026
Generating Less Boilerplate with AI and Local Code Templates
AIR separates AI-generated instructions from template-based code, reporting 45–71% fewer tokens on boilerplate-heavy tasks.
Read more →Papers Added Oct 15, 2026
AI-Generated Code Can Work Without Being Proven Secure
A conceptual report explains false security confidence: when AI-generated code seems to work, but its security is still poorly assessed or understood.
Read more →Papers Added Oct 15, 2026
Can reasoning prompts make AI-generated code more robust?
An automated review examines reasoning prompts and code-generation robustness, but its abstract does not report whether the approach works.
Read more →Papers Added Oct 15, 2026
Training AI to Write Safer Code Without Sacrificing Correctness
A training framework focuses on small code repairs to improve the security of AI-generated code while preserving or improving functional correctness.
Read more →Papers Added Oct 15, 2026
AI Coding Costs Depend More on Repair Rounds Than Code Length
A small experiment found no reliable language cost ranking; repeated repair rounds and accumulated context mattered more than the amount of code generated.
Read more →Papers Added Oct 15, 2026
TreeCoder guides AI code generation with checks beyond the prompt
TreeCoder explores ways to apply constraints while AI generates code, improving accuracy in experiments across Python, SQL and Rust.
Read more →Papers Added Oct 15, 2026
Treat AI-Generated Code as Something to Verify, Not Automatically Trust
A conceptual framework outlines how verification, behavioral checks, and traceability can support trust in AI-generated enterprise software.
Read more →Comparisons Added Aug 25, 2026
Lovable vs Bolt vs Replit: The 2026 Showdown
We built the same app three times to see which AI tool gets you to launch fastest — with honest trade-offs.
Read more →Tool Directory Added Aug 23, 2026
The 25 AI Building Tools Actually Worth Your Time
Our curated directory: app builders, agents, design tools, and databases — with honest one-line verdicts on each.
Read more →Guides Added Aug 22, 2026
12 Things to Check Before You Share Your App
Auth, RLS, custom domain, error pages, analytics — the boring stuff that makes the difference between a demo and a product.
Read more →Guides Added Aug 21, 2026
The Prompting Playbook: 15 Prompts That Actually Work
Copy-paste prompts for building features, fixing bugs, and refactoring — tested across Lovable, Bolt, and Cursor.
Read more →Reports Added Aug 20, 2026
What Vibe Coding Actually Costs: A 6-Month Spending Report
We tracked every dollar spent across AI subscriptions, hosting, and domains for 6 months. The real numbers inside.
Read more →Guides Added Aug 19, 2026
Supabase for Vibe Coders: The Only Database Guide You Need
Tables, RLS, auth, and storage explained for people who have never touched a database before.
Read more →Reports Added Aug 18, 2026
From Vibe Code to Real Business: 5 Case Studies
Five builders who turned weekend AI projects into paying products — what they built, how long it took, and what they'd do differently.
Read more →Reports Added Aug 10, 2026
State of Vibe Coding 2026
What the last year of funding, product launches, and adoption actually tells us about where this space is headed.
Read more →Comparisons Added Aug 6, 2026
Lovable vs Replit vs Bolt: Which Should You Use?
Real differences between the three most popular full-stack AI app builders — not just feature lists.
Read more →Comparisons Added Aug 3, 2026
Cursor vs Windsurf vs GitHub Copilot
Three very different approaches to AI-assisted coding inside an editor — and who each one actually suits.
Read more →Glossary Added Jul 31, 2026
The Vibe Coding Glossary
The handful of terms you'll actually run into, explained without jargon.
Read more →Tool Directory Added Jul 27, 2026
The AI Builder Tool Directory
A quick-reference rundown of the tools this site covers most, and what each one is actually for.
Read more →Guides Added Jul 23, 2026
Database Basics for Non-Technical Builders
The handful of database concepts worth understanding before you start describing tables to an AI builder.
Read more →Guides Added Jul 16, 2026
The VibeBuilder Reading List
A short list of things worth reading if you want to go deeper than a single tutorial.
Read more →Guides Added Jul 11, 2026
The Pre-Launch Shipping Checklist
The unglamorous checks worth doing before you share your app outside your immediate circle.
Read more → Newsletter
Weekly vibe-coding tips — coming soon
We're getting this ready. Leave your email and we'll let you know the moment it launches. No fluff, unsubscribe whenever.