Fellow API

Product Introduction

1. Fellow API is a programmatic interface designed to integrate meeting data and workflows directly into external systems through secure REST endpoints. It enables automated compliance archiving, real-time workflow triggers based on meeting content, dashboard integrations for analytics, and seamless data transfer to large language models (LLMs) for customized outputs. The API serves as a bridge between Fellow’s meeting management platform and third-party tools.
2. The core value lies in its ability to transform raw meeting data into actionable enterprise-grade workflows while maintaining strict compliance standards. It eliminates manual data handling by automating record retention, enabling keyword-triggered actions, and providing structured data outputs for AI/ML processing.

Main Features

1. Workspace Governance provides administrative control over API access, including granular user key management, activity monitoring via last-used timestamps, and workspace-level enablement/disablement of API functions. Admins can revoke keys instantly and audit access patterns through detailed logs.
2. Developer-First Design offers clean REST endpoints with comprehensive documentation, prebuilt code samples in multiple languages, and sandbox environments for rapid proof-of-concept development. The API supports pagination, filtering, and webhook integrations for event-driven architectures.
3. Enterprise-Grade Security ensures all data transfers occur over encrypted HTTPS connections with mandatory TLS 1.2+, while maintaining SOC 2 Type II-compliant 90-day audit logs for all API requests. Compliance teams gain automated e-discovery capabilities and governance tool interoperability without additional configuration.

Problems Solved

1. The API addresses the inefficiency of manual meeting data extraction and processing by automating compliance record retention, real-time workflow triggers, and bulk data transfers to analytics platforms. It solves compliance gaps in meeting documentation retention and audit trail generation.
2. Primary users include DevOps engineers building custom integrations, compliance officers managing data retention policies, and product teams embedding meeting insights into SaaS applications. Security teams benefit from built-in governance controls.
3. Typical scenarios involve automatically archiving meeting transcripts to legal hold systems, triggering Slack alerts when contracts mention specific clauses in real time, feeding OKR metrics from meetings to BI tools like Tableau, and generating AI-powered executive summaries using GPT-4.

Unique Advantages

1. Unlike generic workflow APIs, Fellow API combines meeting-specific data structuring with enterprise access controls, providing domain-optimized endpoints for agendas, action items, and transcripts rather than requiring custom field mapping.
2. The 90-day immutable audit log exceeds standard security offerings by timestamping every API call with user/IP metadata, satisfying legal discovery requirements without additional logging infrastructure. Permission-aware scopes enforce least-privilege access to meeting data.
3. Competitive differentiation stems from native integration with Fellow’s AI meeting transcription stack, enabling direct access to processed NLP outputs like sentiment markers and topic clusters that competitors require users to compute separately.

Frequently Asked Questions (FAQ)

1. Who can enable Fellow’s API? Workspace owners or admins with enterprise permissions can activate the API through granular role-based access controls, while individual contributors require explicit approval through defined authorization workflows.
2. Is access permission-aware? Yes, API keys inherit granular OAuth2-style scopes restricting access to specific data types like read-only transcripts or write-capable action items, aligned with the user’s existing platform permissions.
3. What data format do responses use? All endpoints return JSON-formatted data with standardized schemas containing ISO 8601 timestamps, UTF-8 encoded text fields, and nested objects for meeting participants/action items. Pagination follows RFC 5988 link headers.
4. How is it secured and audited? End-to-end encryption uses AES-256 with perfect forward secrecy, while audit logs record API method types, response codes, and byte counts for 90 days—exceeding standard retention policies. Third-party penetration testing occurs quarterly.
5. Can we integrate custom LLMs? Yes, the API provides raw transcript streams and preprocessed NLP outputs (entity recognition, topic segmentation) compatible with major AI platforms through JSON-LD formatting, enabling direct feeding into private model pipelines.