Blog

Predicate Blog

Insights on AI agents, web automation, and building the perception layer for the next generation of AI applications.

Engineering
February 6, 20268 min read

TDD for AI Agents: Why Probability Needs Determinism

Prompt engineering has hit a ceiling. The next reliability leap comes from runtime verification: assertions, traces, and loud failures.

Read article
Engineering
January 31, 20267 min read

Why Agent Failures Are Governance Failures

Agents don't fail because they reason badly. They fail because they act without accountability. This post explains the architectural pattern that fixes it.

Read article
Engineering
January 30, 202610 min read

Debugging a 'Hard' browser READ task with Predicate traces (APA.org case study)

A technical case study on apa.org showing how verification-first execution + JSONL traces make browser agents debuggable (candidate coverage, no-ops, auth drift, extraction grounding).

Read article
Engineering
January 25, 20268 min read

Debugging a flaky ecommerce agent with traces: Ace Hardware WebBench case study

How AgentRuntime traces + snapshot diagnostics turn 'it closed by itself' into actionable fixes (overlays, verification, typing, read injection).

Read article
Engineering
January 20, 20269 min read

A verification layer for browser agents: Amazon case study

How structured snapshots + Jest-style assertions make small local models reliable.

Read article
Engineering
January 16, 20267 min read

Running Real Browser Agents on a 3B Local Model (No Vision Required)

How structure-first snapshots and assertions cut browser agent token usage by ~50% and make small local models viable.

Read article
Engineering
January 15, 20266 min read

How to Run Web Agents with Local LLM (3B)

Run web agents on 3B–14B local models by replacing screenshots + raw DOM with structure-first snapshots — cutting token usage by ~50% per step.

Read article
Engineering
January 15, 20268 min read

Why Structure Beats Vision (and Raw AX) for LLM Web Agents

Modern LLM agents struggle on the web because the web is not a static document—it's a live, unstable system. This post explains why structure + stability + verification beats raw DOM, accessibility trees, or vision-first approaches.

Read article
Engineering
January 15, 20267 min read

Why AX Alone Isn't Enough for Reliable Agents

The Accessibility Tree (AX) provides standardized roles, names, and states—but reliable agents need more. This post explains where AX shines, where it falls short, and why geometry + stability + verification are required.

Read article
Engineering
January 3, 20265 min read

From Screenshots to Audit Trails: Why Agent Observability Matters

Why logs and screenshots are not enough for AI agents — and how trace-based observability enables replay, determinism, and real debugging.

Read article
Engineering
December 28, 20256 min read

Why Vision-Based Web Agents Fail — and What Semantic Geometry Fixes

Vision models are good at seeing, but agents fail at acting. This post explains why vision-first web agents break down in practice, and how semantic geometry enables reliable execution.

Read article
Product Update
December 20, 20258 min read

Cut Your AI Agent Costs by 95% with Smart Filtering

How we solved the '4,500 elements problem' and saved customers thousands in token costs without sacrificing accuracy.

Read article
Product
December 15, 20255 min read

Introducing Sentience: The Perception Layer for AI Agents

Building AI agents that can truly see and understand the web requires more than just scraping HTML. Learn how Sentience provides visual grounding for large action models.

Read article
Engineering
December 10, 20257 min read

How We Achieved Sub-Second Web Rendering

Traditional headless browsers are slow and expensive. We built an adaptive hybrid architecture that delivers 10x faster performance at 90% lower cost.

Read article
Guide
December 5, 20256 min read

Precision vs Performance: Choosing the Right Rendering Mode

Not all web automation tasks are created equal. Learn when to use our Performance Engine for speed and when to use Precision Engine for accuracy.

Read article

Stay Updated

Get the latest articles on AI agents, web automation, and developer tools delivered to your inbox.