Provides a natural language-to-SQL API that enables developers to query structured databases using plain language, integrating seamlessly into existing data stacks with minimal code. The platform improves data accessibility by converting user queries into accurate SQL commands, supporting fine-tuning and synthetic data generation for enhanced performance. Open source and usage-based pricing options allow for flexible deployment and cost management.
Funding
$3.1M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Many business users lack SQL expertise, making it difficult for them to directly query databases and extract insights from structured data. This reliance on technical staff for data access creates bottlenecks and delays in obtaining critical information for decision-making. Traditional methods of data access often require complex coding and database management skills, hindering self-service analytics.
Solution
Dataherald provides a natural language-to-SQL API that enables users to query structured databases using plain language, eliminating the need for SQL knowledge. The platform integrates into existing data stacks, allowing developers to embed the API into their applications with minimal code. By converting natural language queries into accurate SQL commands, Dataherald improves data accessibility and empowers business users to perform self-service analytics. The engine combines custom agents with fine-tuning and built-in evaluation to deliver accurate text-to-SQL performance.
Target Audience
The primary users are developers and business users who need to query structured databases using natural language, particularly those in SaaS companies and data-driven organizations.
Features
- Natural language processing (NLP) engine that translates user questions into SQL queries
- API that integrates with existing databases and SaaS applications
- Fine-tuning support for GPT 3.5 and GPT 4 models to improve accuracy and latency
- Built-in evaluator to monitor model performance and enable feedback learning
- Synthetic data generation to improve agent performance
- Admin console for configuring and observing queries, models, and fine-tuning
- Usage-based pricing for self-serve users