Appbase

Appbase (appbase.io) develops ReactiveSearch, a stack for building site search that combines composable search workflows, front-end UI components, analytics and access controls. Developers author search pipelines and transformations in JavaScript and can route requests across multiple backends, including Elasticsearch, OpenSearch, OpenAI and generic HTTP-based APIs, via built-in connectors. A configurable cache layer is offered to increase search and AI answer throughput, and fine-grained access control is configurable through Basic Auth and JWT with point-and-click policy setup, including team access policies.

On the interface side, the company publishes low-code UI kits for React, Vue, Flutter and vanilla JavaScript, including the ReactiveSearch components library and ReactiveMaps for data-aware realtime map components. Analytics features cover search, filter and result engagement, popular, recent, featured and FAQ suggestions for relevance enrichment, natural-language Q&A intent capture, and recommendations addressing relevance, performance and content gaps.

The company maintains a broader open-source portfolio around Elasticsearch tooling: dejavu, a web UI for viewing and managing Elasticsearch/appbase.io data; mirage, a GUI for composing Elasticsearch Query DSL; gem, a GUI for data modeling; Docbase for turning Markdown documents into sites; client libraries such as appbase-js for JavaScript/Node.js/React Native and go-appbase for Go; searchbox lightweight search UI libraries; arc, an API gateway for Elasticsearch providing security, rate limiting, analytics and request logging; and ABC, a command-line tool for managing appbase.io accounts and for importing and syncing data into Elasticsearch from Postgres, MySQL, SQL Server, MongoDB, CSV, JSON, Kafka and Elasticsearch itself, with realtime sync currently supported for MongoDB and Postgres.

Founding story

Siddharth Kothari and Sagar Chandarana conceived Appbase while working on backend infrastructure at Mathify, an interactive math learning platform, where they encountered difficulties orchestrating backend systems reliably and attributed the problem to missing abstractions and repeated ad-hoc integrations. Both hold Computer Science degrees from the Dhirubhai Ambani Institute of Information and Communication Technology, and previously worked at Google and Carnegie Mellon University; Kothari completed his degree in 2013. The founders had known each other for over four years at launch, and the company's first $10,000 check came from the entrepreneurship cell at DA-IICT.

Business model

Appbase distributes its core ReactiveSearch stack as free and open-source software that customers can self-host with Docker, with support provided through the GitHub community and enterprise support offered on request. Historically the company also operated appbase.io as a hosted, managed database and search service with user accounts, apps and metrics, and maintained a proprietary variant of its ABC tooling alongside the open-source base project, including a closed-source import module.

The core software is free and open source and self-hostable; enterprise support is available on request. The earlier appbase.io service was a managed, account-based hosted database and search offering with per-app provisioning and metrics, and the company maintained a proprietary build variant of its ABC tooling.

Traction

The company reports more than 15,000 GitHub stars, over 5 million downloads and over 300 million searches per month across its stack, with the UI libraries alone accounting for over 5,000 stars and more than 1 million downloads. Individual repositories show 471 stars for the ABC CLI and 42 stars for appbase-js. In its earliest phase, over 130 apps were built on the platform during the September beta and more than 4.8 million API calls had been served by February 2015.

Latest developments

The ReactiveSearch stack is now offered as free and open source, positioned around incremental adoption of AI for site search, with connectors that unify Elasticsearch, OpenSearch, OpenAI and HTTP APIs in a single composable workflow, plus a caching layer for search and AI answer throughput and point-and-click access controls. The company reports over 15,000 GitHub stars, more than 5 million downloads and over 300 million searches per month across the stack.