Always-On AI Agents

Agents that continuously model, monitor, and analyze your data — surfacing anomalies, trends, and insights proactively, before anyone asks.

Expose Altertable as an MCP server so your own agents can use typed tools to retrieve governed data, reason over it, and trigger scoped actions — with sub-second latency and production-grade controls. Every insight gets captured as organizational memory, so context accumulates over time instead of disappearing into Slack.

Capabilities

Core features that enable agents to query and act on your data

MCP Server

Make use of Altertable's MCP server. Agents connect via a standard protocol to discover tools, query data, and act as human workers.

Everything Is a Tool

Every query, insight, segment, dashboard and action is a resource or a tool. Agents can discover and use these tools to build workflows.

Additional Capabilities

More features that make Altertable a complete agentic workflows platform

Query and Act

Agents can both retrieve insights and trigger actions—running queries, generating segments, and writing back results where permitted. Database docs.

Sub Second Context

Backed by our analytical engine and adaptive caching, agents receive sub-second answers—enabling tight perceive-decide-act loops.

Bring Your Own Agentic Framework

Bring your own preferred agentic framework. Connect via MCP to leverage Altertable's stored data and query capabilities.

Every language. Every stack.

SDKs and adapters for all major languages, so your team ships without friction.

Lakehouse API

Query and write data directly to your Altertable lakehouse.

from altertable_lakehouse import Client
from altertable_lakehouse.models import QueryRequest

client = Client(
    username="your_username",
    password="your_password",
)

# Run a SQL query
req = QueryRequest(statement="""
    SELECT
        date_trunc('day', timestamp) AS day,
        COUNT(DISTINCT user_id)      AS dau
    FROM   events
    WHERE  event = 'Event Name'
    AND    timestamp >= NOW() - INTERVAL '30 days'
    GROUP  BY 1
    ORDER  BY 1 DESC
""")

# Accumulate all rows in memory
result = client.query_all(req)
for row in result.rows:
    print(row)

# Append data to a table
client.append(
    catalog="my_catalog",
    schema="my_schema",
    table="users",
    data={"user_id": 1, "plan": "pro"}
)