API Changelog

A new interactive playground lets you craft requests, inspect responses, and explore every endpoint directly from the developer portal. Includes request history and built-in auth handling.

Added comprehensive Quick Start guide with step-by-step instructions, and a Concepts page explaining optimization models, time windows, vehicle capacities, and constraint types.

You are looking at it! A searchable, filterable changelog to keep you informed of every API change.

Optimization results now include encoded polyline geometry for every route leg, enabling direct map rendering without a separate directions call.

Aligned MlModelStatus, MlPerformanceResponse, DemandForecastResponse and MlRetrainResponse schemas with actual service return values. Removed speculative fields that were never populated.

Fixed a mismatch between the billing summary Pydantic schema and the actual database query results. The response now correctly includes usage breakdown by plan tier.

Resolved an issue where job listing could return results in non-deterministic order when multiple jobs shared the same creation timestamp.

Introduced an AI-powered chat assistant that understands your optimization context. Supports multi-turn conversations with context awareness across jobs, routes, and analytics.

New LLM-powered endpoints can explain optimization results in plain English, detect anomalies in route patterns, and parse natural language scheduling requests into structured optimization payloads.

Fixed AnalyticsRoutesResponse, AnalyticsCarbonFootprint, AnalyticsEfficiencyTrends, and AnalyticsRecommendation schemas to match actual service output. Removed ghost fields that returned null.

Register webhook URLs to receive real-time notifications for job completion, failures, and status changes. Includes delivery logs, automatic retries with exponential back-off, and signature verification.

Schedule recurring PDF and CSV reports for fleet analytics, cost breakdowns, and SLA compliance. Reports are generated asynchronously and delivered via webhook or email.

New endpoints for GDPR compliance allow users to export all personal data as a downloadable archive and request permanent account deletion with a 30-day grace period.

Submit partial changes (new tasks, removed vehicles, shifted time windows) to an existing solution and receive an updated plan without re-solving from scratch.

Compute travel time and distance matrices for car, truck, and bicycle profiles. Supports up to 100x100 origins/destinations per request using OSRM.

Machine learning models predict travel times based on historical traffic patterns. Trigger retraining with your own fleet data for improved accuracy.

All API actions are now recorded in an immutable audit log. Organization admins can invite team members with role-based access.

Balance competing objectives like total distance, number of vehicles, time window violations, and task priority using configurable weights in the optimization request.