🩺 Vitals
- 🟢 Last active: 2026-08-04
- 📦 Latest release: v0.0.35 (2026-07-25)
- 🐞 Open issues: 117
- 🌟 Stars: 15.8k
What do these metrics mean?
- Last active: when code was last pushed, as of our last check. The dot is green when that was recent, grey otherwise. A long gap can mean a tool is finished and stable, not only unmaintained.
- Latest release: the most recent tagged, packaged version the maintainers published. Not every healthy project tags releases.
- Open issues: unresolved reports and requests. A high number is normal for a popular project and is not a warning on its own.
- Stars: how many people bookmarked the project on its forge. A rough popularity signal, not a measure of quality.
🏗️ Profile
- Official: surfsense.com
- Source: github.com/MODSetter/SurfSense
- License: Apache-2.0
- Deployment: Docker
- Data Model: PostgreSQL + PGVector + Neo4j (Graph)
- Jurisdiction: USA 🇺🇸 (SurfSense Inc.)
- Compliance (SaaS): N/A (No certifications published)
- Compliance (Self-Hosted): Self-Hosted (User Managed)
- Complexity: Medium (3/5) - Multi-service stack (PostgreSQL with PGVector, Neo4j, LLM orchestration); Docker Compose required
- Maintenance: Medium (3/5) - Actively developed; startup team; no formal release cadence established
- Enterprise Ready: Low (2/5) - No SSO, RBAC, or compliance certifications; designed for internal self-hosted deployment
1. The Executive Summary
What is it? SurfSense is an open-source private AI search engine that connects local LLMs to internal knowledge bases, indexing content from Slack, Notion, Jira, Google Drive, and browser research into a vector and graph database for cited, sovereign AI search. Developed by SurfSense Inc. (USA) and positioned as a self-hostable alternative to Google NotebookLM and Perplexity, it is optimised for air-gapped deployment where proprietary SaaS search tools create unacceptable data residency risk. The Apache-2.0 core is fully functional; the project is actively transitioning toward a commercial model with enterprise features to be introduced in future releases.
The Strategic Verdict:
- 🔴 For Regulated Industries or Teams Requiring Compliance Attestations: Caution. SurfSense publishes no SOC 2, ISO 27001, or GDPR documentation. Self-hosting is mandatory, and even on self-hosted, the operator absorbs the full security posture. The SaaS offering is not suitable for corporate data ingestion.
- 🟢 For Internal R&D and Engineering Teams: Strong Buy. When deployed within a secure VPC, SurfSense delivers sovereign AI search over internal knowledge bases without routing sensitive document content through vendor infrastructure. Apache-2.0 licence ensures flexibility and long-term ownership of the search pipeline, for as long as the licence holds.
2. The "Hidden" Costs (TCO Analysis)
| Cost Component | Google NotebookLM (SaaS) | SurfSense (Self-Hosted) |
|---|---|---|
| Data Privacy Risk | High (Google cloud ingestion) | Zero (air-gapped capable) |
| LLM Vendor Lock-in | Google Gemini only | Any LLM provider |
| Connector Ecosystem | Google Workspace only | Slack, Notion, Jira, Drive |
| Compliance Posture | Google's certifications | Operator-managed |
3. The "Day 2" Reality Check
🚀 Deployment & Operations
- Installation: Deployed via Docker Compose, orchestrating the SurfSense application, PostgreSQL with PGVector extension, and a Neo4j graph database. An LLM provider (Ollama for local inference or a remote API) must be configured separately.
- Scalability: Suited for individual or small-team knowledge hubs. Enterprise-scale deployment across large document corpora and concurrent users remains in active development; production readiness at scale should be validated before committing to critical search infrastructure.
🛡️ Security & Governance (Risk Assessment)
- Jurisdiction & Geopolitics (USA 🇺🇸): SurfSense Inc. is a US-incorporated startup subject to the CLOUD Act. The architecture is optimised for fully local, air-gapped deployment. Self-hosting eliminates CLOUD Act exposure entirely by ensuring no data transits vendor infrastructure. EU operators should treat any SaaS-hosted offering as non-compliant until formal data processing documentation is published; no GDPR Data Processing Agreement exists at the vendor level.
- The Compliance Shift: SurfSense publishes no compliance certifications: no SOC 2, ISO 27001, or GDPR documentation exists at the vendor level. Self-hosting is the only viable path for regulated enterprise use. The operator absorbs the full compliance posture: PostgreSQL and PGVector security, Neo4j access controls, LLM API key management, and connector token storage are entirely the operator's responsibility. SaaS connector integrations (Slack, Notion, Jira) require persistent read-access tokens stored in the operator's database. This layer requires rigorous internal security review before production deployment.
- License Risk (Apache-2.0: Permissive Now; Commercial Transition Risk): The current codebase is Apache-2.0 licensed, permissive and forkable. However, SurfSense is openly transitioning from a community OSS project to a commercial startup, with a published commercialisation roadmap and active hiring. This trajectory carries a medium-high risk of future licence changes (BUSL or SSPL) as enterprise features are introduced and investor pressure mounts. Organisations deploying SurfSense on critical internal search infrastructure should pin to a verified Apache-2.0 release and maintain a fork strategy as a contingency against relicensing.
4. Market Landscape
🏢 Proprietary Incumbents
- Google NotebookLM: A powerful AI synthesis tool locked to the Google ecosystem. All document content is ingested into Google's cloud infrastructure, a data residency and training-data risk for enterprises handling sensitive internal knowledge.
- Glean: An enterprise-grade AI search platform with deep SSO and security integration. High procurement cost and a fully managed SaaS architecture with no self-hosting option.
🤝 Open Source Ecosystem
- AnythingLLM: A local RAG application covering similar private AI search use cases: broader document source support with a single-container deployment model, preferred by teams prioritising ease of setup over SaaS connector depth.