AI Workflows & Agentic Architecture Directory

Google dominates search, but AI-powered alternatives are growing fast. This deep dive explores Perplexity, ChatGPT Search, and the fundamental shift from ten blue links to AI-generated answers.

AI agents can now operate desktop applications through visual grounding and GUI action tokens. This workflow builds gui-agent, a LangGraph pipeline that takes a natural-language task, captures screenshots, identifies UI elements with visual grounding, generates click/type/scroll actions, and executes multi-step workflows across desktop applications.

Cloud AI has latency and privacy costs. Edge AI runs models directly on devices. This workflow builds edge-infer, a LangGraph pipeline that routes inference requests to quantized models running on-device via WebGPU, with adaptive model selection based on task complexity and device capabilities.

Documents are not just text — they contain images, charts, tables, and diagrams. This workflow builds multi-rag, a LangGraph pipeline that ingests multi-modal documents, indexes both text and visual content, and retrieves information across modalities using vision-language models and hybrid search.

Computer-use agents need governed tools for GUI interaction. This MCP server gives AI agents the ability to capture screenshots, detect UI elements, and execute click/type/scroll actions on any visible application.

AI agents produce unstructured text that downstream systems cannot parse reliably. This MCP server gives agents the tools to validate, format, and enforce structured output against JSON schemas before returning results.

The AI industry obsessed over bigger models. But in 2026, small language models under 3B parameters are outperforming 100B+ giants on real production tasks.

AI coding agents are not free. This deep dive breaks down the unit economics.

Language models have learned to understand the world through text. Now they are learning to act in it.

Most AI governance frameworks are policy documents that sit on shelves.

Enterprise finance teams process millions of transactions daily, and manual reconciliation is a top source of accounting errors. This workflow builds recon-fleet, a LangGraph pipeline with Temporal durable execution that matches transactions across ledgers, flags discrepancies, routes disputes through human review, and produces an audit trail — turning a days-long manual process into an automated, auditable workflow.

95% of clinical trials miss their enrollment targets, and the primary bottleneck is patient-to-trial matching. This workflow builds trial-match, a LangGraph pipeline that ingests de-identified patient profiles, matches them against active trial criteria, scores eligibility, and routes borderline cases through a clinician escalation gate — all within a privacy-preserving architecture that never exposes raw patient data to the model.