Building Agents

Design RAG pipelines and measure retrieval with precision, recall and NDCG

Score AI work on two axes and catch inflated self-grades before you trust them

Turn a build sheet into API payloads and a resumable script to launch the agent

Recap every piece of a launched managed agent and suggest the next upgrades

Look up current OpenAI docs with citations and pick the right model for a use case

Answer AI SDK questions and help build chatbots, agents and RAG features

Call Gemini web for text, images and multi-turn chats through a reverse-engineered API

Browse API cookbook examples and tutorials for building LLM-powered apps

Measure and improve model and agent quality with eval datasets and failure analysis

Run inference with Gemini and third-party open models on Agent Platform

Manage RAG corpora and retrieve grounded contexts with the Google GenAI SDK

Meter LLM token use and screen prompts with an Apigee proxy in front of agents

Build AI flows and agents in Dart and Flutter apps with the Genkit SDK

Build AI features, flows and tools in Go with the Genkit framework

Build AI agents, flows and tools in TypeScript with the Genkit framework

Build AI agents, flows and tools in Python with the Genkit framework

Create and manage stateful server-side Agent resources with files, skills and tools

Call the Gemini API on Google Cloud with the Gen AI SDK in several languages

Use the Gemini Interactions API for multi-turn, streaming and background runs

Generate a Gemini Live API websocket client with session handling and token refresh

Create, test and deploy ADK agents on Google Cloud with agents-cli

Design live multi-agent systems on Google Cloud for real-time streaming data

Design, build and deploy multi-agent systems on Google Cloud

Secure multi-agent traffic with Agent Gateway, Model Armor and IAP

Design an enterprise RAG search system on GKE with a SQL vector store

Develop and operate Agent Zero projects, tasks, tools and its API and WebUI

Write DeepEval evals and datasets to test AI agents, RAG pipelines and chatbots

Export OpenTelemetry traces from LLM apps and agents to Confident AI

Add DeepEval tracing with @observe to LLM apps and agents for Confident AI

Build LLM-as-judge evals with pairwise comparison, rubrics and bias checks

Model agent beliefs, desires and intentions with BDI concepts and RDF transforms

Build agent evaluation suites with regression tests, rubrics and quality gates

Decide whether an LLM fits a task and estimate token costs for pipelines

Debug agent runs by reading reasoning traces for drift, tool confusion and regressions

Run a local voice assistant that explains and diagrams your codebase

Benchmark and tune A2UI inference formats for accuracy, latency and token cost

Build a local BM25 search index over your vault with optional semantic reranking

Build chat, streaming and tool calling with Vercel AI SDK 5 and migrate from v4

Add realtime speech to speech voice conversations to an app with Grok

Build apps, scripts and CI automations on the Cursor TypeScript SDK

Capture and inspect realtime voice session logs to debug voice assistant issues

Answer Cursor Origin API questions and check the key rules for Origin Apps

Plan the port of a GitHub App to a Cursor Origin App with a porting brief

Evaluate code generation models on HumanEval, MBPP and other benchmarks with pass at k

Extract structured data from LLM responses with Pydantic validation and automatic retries

Build LLM apps with LangChain agents, chains, tool calling and RAG retrieval

Trace, evaluate and monitor LLM applications with LangSmith datasets and tests

Build RAG apps with LlamaIndex document ingestion, indexes and query engines

Run LLM benchmark suites with 100+ tasks on Docker, Slurm HPC or cloud platforms