AI visibility report for Sourcegraph
Vertical: Version Control & Code Collaboration
AI search visibility benchmark across 5 platforms in Version Control & Code Collaboration.
Also benchmarked
Sourcegraph appears in another vertical
Presence Rate
Top-3 citations across 125 prompt × platform pairs
Sentiment
Peer Ranking
Key Metrics
Platform Breakdown
Grok
Perplexity
Gemini Search
Google AI Mode
ChatGPTOverview
Sourcegraph is a San Francisco-based code intelligence platform founded in 2013 by Quinn Slack and Beyang Liu. Designed to address 'Big Code'—the challenge of managing large, complex, multi-repository codebases—Sourcegraph provides enterprise engineering teams and AI agents with unified search, navigation, and understanding across all code regardless of host. Core capabilities include exhaustive cross-repository Code Search, agentic Deep Search, Batch Changes for large-scale automated edits, Code Insights for analytics, Code Monitoring, and an MCP server for AI agent context. The company also launched Amp, an independent AI coding agent, in 2025. Trusted by 200+ enterprise teams including Stripe, Reddit, BlackRock, and Nutanix, Sourcegraph has indexed over 54 billion lines of code. It has raised $223M at a $2.6B valuation (Series D, 2021).
Sourcegraph is a code intelligence platform that indexes all of an organization's repositories across every code host, enabling developers and AI agents to search, understand, and evolve large, complex codebases. Its platform spans universal code search (exact, regex, structural, and diff search), semantic code navigation, agentic natural-language Deep Search, automated large-scale Batch Changes, Code Insights dashboards, real-time Code Monitoring, and an MCP server that feeds codebase context to external coding agents. Cody, an AI coding assistant, remains available for enterprise customers. The platform is deployed as single-tenant cloud or self-hosted and is built for enterprise security and scale with SOC 2 Type II, ISO 27001, and SSO/SCIM support.
Key Facts
- Founded
- 2013
- HQ
- San Francisco, CA, USA
- Founders
- Quinn Slack, Beyang Liu
- Employees
- 150-200
- Funding
- $223M
- ARR
- ~$50M
- Customers
- 200+ enterprise engineering teams
- Valuation
- $2.6B
- Status
- Private
Target users
Key Capabilities9
- Universal cross-repository, cross-host code search (regex, structural, literal, diff, commit)
- Deep Search: agentic natural-language queries with cited, grounded answers across the codebase
- SCIP-powered semantic code navigation (jump-to-definition, find references, hover docs) across 30+ languages
- Batch Changes: automated large-scale code modifications across all repositories via declarative specs
- Code Insights: queryable dashboards tracking migrations, dependency versions, vulnerability remediation, and code trends over time
- Code Monitoring: real-time alerts to Slack, PagerDuty, Jira, or webhooks when code patterns change
- MCP server for supplying full codebase context to external AI coding agents
- Living Documentation: auto-generated, continuously updated knowledge base from the codebase
- SOC 2 Type II + ISO 27001 compliance with zero LLM data retention and RBAC/SCIM for enterprise security
Key Use Cases8
- Cross-repository code search and navigation in large, multi-host enterprise codebases
- Security vulnerability identification and remediation at scale (e.g., Log4j, CVE sweeps)
- Large-scale refactoring, migrations, and dependency upgrades across hundreds of repositories
- Developer onboarding and codebase comprehension acceleration
- Incident response: locating root-cause code and error-message origins across all services
- Providing full codebase context to AI coding agents and LLM-based workflows via MCP
- Engineering analytics: tracking migration progress, tech debt, and adoption metrics
- Compliance and audit: monitoring code changes for security and regulatory patterns
Sourcegraph customer outcomes
4 days to full customer patch delivery; 100% of Log4j instances identified; individual fix queries completed in under 5
Used Sourcegraph to identify every instance of the Log4j vulnerability across a sprawling multi-system codebase and delivered customer patches within four days of the disclosure.
80% reduction in time for large-scale code changes
Replaced an internally maintained automation tool with Sourcegraph Batch Changes to propagate large-scale code changes across 70+ repositories, including React upgrades, Kubernetes API version updates, and CDN migrations.
Up to 40% developer productivity increase for 2,000 developers
Deployed Sourcegraph in a secure environment alongside AWS and Anthropic AI services to boost developer productivity for a team of 2,000 engineers.
Task that would have taken months completed in days by a single engineer
Used Sourcegraph's multi-repository code search to analyze thousands of repositories for GDPR compliance refactoring, a task that would have taken months if done manually per repository.
Connected Sourcegraph's MCP server to Stripe's internal 'Minions' AI coding agents to supply code intelligence context, enabling agents to gather internal docs, ticket details, build statuses, and code context in a single pipeline.
Recent Trend
How AI describes Sourcegraph3
### Sourcegraph Often considered the benchmark for large-scale code search.
What are the best tools for cross-repository code search at scale — finding all usages of a deprecated function across 300 repos instantly?
| | Sourcegraph Code Search Platform (paired with Git host) | Excellent for search | Not a Git host itself, but often deployed alongside GitHub/GitLab for very large codebases.
Which Git hosting platforms handle massive monorepos best — which ones stay performant when a repo grows to millions of files and hundreds of thousands of commits?
The Gold Standard: Sourcegraph ---------------------------------- If your primary goal is to search across hundreds or thousands of repositories instantly with 100% precision, Sourcegraph remains the market leader.
What are the best tools for cross-repository code search at scale — finding all usages of a deprecated function across 300 repos instantly?
Most cited sources8
- 15
Multi-repo search: How to search across multiple repositories | Sourcegraph
sourcegraph.com·Blog Post
- 15
Why code search at scale is essential when you grow beyond one repository | Sourcegraph
sourcegraph.com·Blog Post
- 10
Code Search | Sourcegraph
sourcegraph.com·Blog Post
- 6
AI Coding Context Tools Compared: Agents, Editors, MCPs & Sourcegraph | Sourcegraph
sourcegraph.com·Blog Post
- 5
13 Best Automated Code Review Tools in 2026: AI and Static Analysis Compared | Sourcegraph
sourcegraph.com·Blog Post
- D4
GitHub code search vs. Sourcegraph - Sourcegraph docs
docs.sourcegraph.com·Documentation
Alternatives in Version Control & Code Collaboration6
Sourcegraph occupies a distinct niche as a code intelligence and universal code search layer that sits atop existing version control systems rather than replacing them.
- Unlike GitHub, GitLab, or Bitbucket—which bundle code search within a broader SCM and DevOps platform—Sourcegraph is purpose-built to index, search, and understand code across all code hosts simultaneously.
- Its differentiator is 'Big Code' capability: exhaustive, cross-repository, multi-host search with SCIP-powered semantic indexing that rivals native GitHub or GitLab search at orders of magnitude larger scale.
- In the AI era, Sourcegraph's MCP server and Deep Search position it as a codebase-context layer for coding agents, competing indirectly with GitHub Copilot's context features but in a host-agnostic, enterprise-grade way.
- The 2025 spin-off of Amp as an independent AI coding agent company narrows Sourcegraph's focus purely to code understanding, oversight, and codebase evolution.
Reviews
Praised
- Exhaustive, deterministic cross-repository code search
- Powerful regex and structural search syntax
- Effective for security vulnerability sweeps across large codebases
- Significant time savings for large-scale refactors via Batch Changes
- Faster developer onboarding in complex codebases
- Strong enterprise security and compliance posture (SOC 2, ISO 27001)
- MCP server enabling AI agent codebase context
Criticized
- High per-user pricing compared to native GitHub or GitLab search
- Discontinuation of Cody Free, Pro, and Enterprise Starter plans in 2025
- Source code made private in 2024, removing open-source transparency
- Frequent strategic pivots creating product inconsistency
- Self-hosted deployment complexity and maintenance overhead
- Leadership churn and organizational instability noted by employees
- Limited native integration with non-GitHub/GitLab code hosts
User sentiment toward Sourcegraph is generally positive for its core code search capability, which is widely praised as exhaustive, fast, and reliable across large multi-repository environments—a key differentiator versus native GitHub or GitLab search. Enterprise customers highlight the value for security vulnerability remediation, large-scale refactors, and developer onboarding. Criticisms center on high pricing relative to native alternatives, product strategy volatility (multiple pivots over 2022–2024 noted by both users and employees), the 2023–2024 transition away from open source, and the discontinuation of lower-tier Cody plans in 2025. Internal Glassdoor and Blind reviews reflect organizational challenges and leadership churn, though technical quality of the core product is consistently respected.
Pricing
Sourcegraph's primary commercial offering is Enterprise Search at $49 per user per month (single-tenant cloud, billed annually), covering Code Search, Deep Search, Batch Changes, Code Insights, Code Navigation, Code Monitoring, and 24×5 support with optional premium tiers. An Enterprise Starter plan at $19/user/month (up to 50 developers, up to 100 repositories) was introduced in early 2025 but had Cody features removed from it by June 2025 following the discontinuation of Cody Free/Pro plans. Self-hosted (on-premises) enterprise pricing is negotiated based on user count and required features. Deep Search usage is included at 3 queries/user/month for Code Search seats, with additional usage charged incrementally. Amp, spun out as a separate entity, uses a credit-based pricing model. Observed Vendr transaction values range from ~$15K/year for small teams to $250K+/year for large enterprise deployments.
Limitations
- Source code was relicensed from Apache 2.0 to a proprietary enterprise license in 2023 and made fully private in August 2024, removing community transparency and hackability.
- Cody Free, Cody Pro, and Enterprise Starter plans were discontinued effective July 2025, narrowing self-serve access.
- Pricing is enterprise-oriented and can be cost-prohibitive for small teams compared to native code search in GitHub or GitLab.
- Self-hosted deployments require significant infrastructure and ongoing engineering overhead.
- The 2025 spin-off of Amp as a separate company introduced organizational uncertainty.
- Multiple employee reviews cite a pattern of strategic pivots and leadership churn, which has affected product consistency.
- Native integrations do not extend to all code hosts; some require workarounds using the Src-srv-git API.
Frequently asked questions
Topic Coverage
Prompt-Level Results
| Prompt | Gemini Search | Perplexity | ChatGPT | Grok | Google AI Mode |
|---|---|---|---|---|---|
Capability1/5 cited (20%) | |||||
What are the best tools for cross-repository code search at scale — finding all usages of a deprecated function across 300 repos instantly? | |||||
Which code review tools offer the most useful AI suggestions — which platforms surface actionable feedback rather than just adding noise for reviewers? | |||||
Which code collaboration platforms handle large binary files and machine learning model artifacts best alongside source code in the same repo? | |||||
Which platforms offer the best pull request automation — auto-merging, required checks, dependency updates — without introducing excessive risk? | |||||
Which enterprise Git platforms offer the most flexible access control — which ones support read-only contractor access to specific repos without a full seat license? | |||||
Developer Experience3/5 cited (60%) | |||||
Which tooling makes stacked pull request workflows less painful to manage in practice — what do teams actually use for this? | |||||
What tooling helps reduce merge conflicts in a monorepo where dozens of engineers are committing to overlapping areas simultaneously? | |||||
What engineering metrics platforms give managers visibility into cycle time and review turnaround without micromanaging the team? | |||||
Which code review platforms are best at keeping review workflows fast and avoiding multi-day bottlenecks — what features actually make the difference? | |||||
What tools help large engineering teams manage code ownership so the right reviewers are auto-assigned without spamming everyone? | |||||
Integrations & Ecosystem0/5 cited (0%) | |||||
Which code review platforms have the best CI/CD pipeline integrations for making failed checks automatically block merges? | |||||
Which enterprise code collaboration platforms offer the best SAML/SSO and SCIM provisioning support for automated user lifecycle management? | |||||
Which code collaboration platforms integrate most deeply with project management tools — which ones can automatically move a ticket to in-review when a commit references it? | |||||
Which code collaboration platforms integrate best with secrets scanning and security scanning tools to block vulnerable code before it merges? | |||||
Which Git platforms offer the best webhook and API support for building internal developer tooling on top of repo events? | |||||
Performance & Reliability0/5 cited (0%) | |||||
Which Git platforms offer the best clone and fetch performance for large repositories — how do cloud-hosted and self-hosted options compare? | |||||
Which hosted Git platforms have the best disaster recovery and geo-redundancy — how long until push and pull are restored after a datacenter failure? | |||||
Which code collaboration platforms offer the best enterprise SLAs and availability during planned maintenance — what should teams compare? | |||||
Which Git platforms offer the best partial clone and sparse checkout support for teams who can't clone an entire massive monorepo locally? | |||||
Which Git hosting platforms handle massive monorepos best — which ones stay performant when a repo grows to millions of files and hundreds of thousands of commits? | |||||
Setup & First Run0/5 cited (0%) | |||||
What tools make it easiest to migrate a 10-year-old SVN repository to Git without losing commit history or tags? | |||||
Which self-hosted Git platforms are best for enterprises with strict data residency requirements — is the operational overhead worth it compared to managed options? | |||||
What monorepo tooling should a 50-person engineering team evaluate — which platforms best support monorepo, polyrepo, or hybrid repository structures? | |||||
What are the best code collaboration platforms for onboarding a remote team of 20 engineers with minimal disruption to active sprints? | |||||
Which code collaboration platforms make it easiest to set up branch protection rules and merge policies from day one for a growing engineering team? | |||||
Strengths2
What are the best tools for cross-repository code search at scale — finding all usages of a deprecated function across 300 repos instantly?
Avg # 1.3 · 3 platforms
What tools help large engineering teams manage code ownership so the right reviewers are auto-assigned without spamming everyone?
Avg # 2.0 · 2 platforms
Gaps5
Which hosted Git platforms have the best disaster recovery and geo-redundancy — how long until push and pull are restored after a datacenter failure?
Competitors on 5 platforms
Which Git platforms offer the best partial clone and sparse checkout support for teams who can't clone an entire massive monorepo locally?
Competitors on 4 platforms
Which Git platforms offer the best webhook and API support for building internal developer tooling on top of repo events?
Competitors on 4 platforms
Which code review platforms have the best CI/CD pipeline integrations for making failed checks automatically block merges?
Competitors on 3 platforms
Which code collaboration platforms integrate most deeply with project management tools — which ones can automatically move a ticket to in-review when a commit references it?
Competitors on 3 platforms
Vertical Ranking
| # | Brand | PresencePres. | Share of VoiceSoV | DocsDocs | BlogBlog | MentionsMent. | Avg PosPos | Sentiment |
|---|---|---|---|---|---|---|---|---|
| 1 | GitHub | 49.6% | 48.2% | 22.4% | 14.4% | 48.8% | #26.0 | +0.19 |
| 2 | GitLab | 39.2% | 30.1% | 28.0% | 20.8% | 39.2% | #21.9 | +0.19 |
| 3 | Bitbucket | 25.6% | 10.9% | 6.4% | 4.0% | 25.6% | #28.5 | +0.21 |
| 4 | Azure DevOps (Microsoft product) | 12.0% | 4.4% | 8.8% | 1.6% | 12.0% | #27.3 | +0.25 |
| 5 | Sourcegraph | 5.6% | 3.2% | 0.0% | 5.6% | 5.6% | #16.8 | +0.51 |
| 6 | Gitea | 5.6% | 2.7% | 4.8% | 0.0% | 5.6% | #24.6 | +0.04 |
| 7 | Graphite | 2.4% | 0.6% | 0.0% | 0.0% | 2.4% | #19.7 | +0.40 |
Turn this into your team dashboard
Sign up to unlock project-level analytics, daily tracking, actionable insights, custom prompt configurations, adoption tracking, AI traffic analytics and more.