​
Agent Engineering

We see agents as the intersection of an LLM, storage, and orchestration. Pixelagent unifies this interface into a single declarative framework, making it the de-facto choice for engineers to build custom agentic applications with build-your-own functionality for memory, tool-calling, and advanced patterns.

Data Orchestration and Storage

Automated persistence, state management, and observability

Tool Integration

Seamless integration with external LLM APIs

Agentic Loops

Built-in support for reflection and planning patterns

Declarative Observability

Easily access your agents memory, tools, and conversation history

Choose Pixelagent when you need robust data management and persistence in your agent systems. The framework is particularly powerful when building agents that need to maintain context, use tools, or handle multiple types of data.

​
Installation and Setup

pip install pixelagent anthropic

from pixelagent.anthropic import Agent

agent = Agent(
    agent_name="financial_analyst",
    system_prompt="You are a CFA working at a top-tier investment bank.",
    model="claude-3-7-sonnet-latest"
)

agent.chat("Explain EPS and how it is calculated.")

agent.chat("What was my first question?")

​
Building Your Agent

​
Core Components

Data Orchestration

Built on Pixeltable’s infrastructure for seamless data management and persistence

Multimodal Support

Built-in support for text, images, and beyond

Declarative Model

Define tables and columns; Pixeltable handles the rest

LLM Integration

Handles OpenAI and Anthropic message protocols

​
Build your own agent

See how we build an Agent with Anthropic

Learn how we craft an agent using Claude, with cost-saving tricks like skipping chat history in tool calls

See how we build an Agent with OpenAI

See how we use GPT models to create a lean, powerful agent with Pixeltable-driven efficiency

​
Extensions and Features

Getting Started

Step-by-step introduction to core concepts

Tool Calling

Learn to integrate external tools and APIs

Reflection

Implement self-improvement loops

Memory Systems

Build long-term memory with semantic search capabilities

Planning

Add sophisticated planning and reasoning loops

Team workflows

Build teams of agents to handle complex tasks

​
Why Choose Pixelagent?

Data-First

Focus on robust data management and persistence

Engineering Freedom

Build exactly what you need without framework constraints

Simplified Workflow

Automated handling of persistence, protocols, and integrations

Ready to start building? Dive into the blueprints, tweak them to your needs, and let Pixelagent handle the infrastructure while you focus on innovation!

​
Getting Help

GitHub

Explore the source code and examples

Issues

Report bugs and request features

Discussions

Join the community discussions

Examples

Browse example implementations

We welcome contributions! Whether it’s adding new features, improving documentation, or sharing example implementations, check out our contribution guidelines to get started.

Was this page helpful?