Baz
Also known as: Baz Technologies
Agentic coding platform whose AI agents review, govern, and secure code across the development lifecycle, from a benchmark leading code review agent and a Figma and Jira spec validator to Baz Planner's pre code gateway that catches risky plans before code is written.
Baz, based in Boston, is an agentic coding platform whose AI agents operate directly on a team's codebase to review, govern, and secure the code that developers and AI copilots produce. Founded by a team of former Palo Alto Networks engineers who helped build its cloud application security business, and led by chief executive Guy Eisenkot, who earlier co founded Bridgecrew before its acquisition by Palo Alto Networks, it raised a seed round that it extended to seventeen million dollars in June 2026, co led by Battery Ventures and Boldstart Ventures. Its premise is that in a world where AI generates code faster than any team can review it, reviewing after the fact is no longer a viable primary defense.
Baz began with an AI Code Review agent that runs on every pull request, deeply understanding the codebase, detecting breaking changes across endpoints and behavior, and enforcing standards across product, design, architecture, security, and reliability, and it ranks first on the precision weighted Code Review Bench. Its Spec Review agent goes further, pulling requirements from Figma over MCP and Jira over REST and using secure browser automation to validate that the delivered experience matches intent, reducing bugs and time to merge. Its newest release, Baz Planner, moves review upstream to a pre code gateway where four agents, a spec reviewer, an advanced security agent, a site reliability engineer that correlates changes against production telemetry, and a fixer that applies and validates safe changes in an isolated runtime, run in a loop to catch and rewrite risky plans before code is written. Throughout, Baz keeps a developer in the loop for every significant decision, accumulates learnings from a team automatically and updates its own instructions, and is built to be observable, explainable, predictable, and reproducible.
Baz reflects its founders' security background, framing itself as AI code governance that gives engineering and security leaders control over a codebase increasingly authored by systems they cannot fully audit, and acting as a super harness that coordinates coding agents rather than replacing them. It runs in the cloud and on Bedrock foundation models rather than offering customer model choice, and it does not yet document a self hosted option. For a team shipping with AI copilots that wants deep, learning code review, spec validation, and pre code security governance integrated into GitHub and the IDE, Baz is a strong and unusually capable fit; a team wanting an on premise deployment, or simply a lightweight linting bot, will find it a governance layer rather than a checkbox.
Vendor details
Canonical URL
https://www.baz.co
Category
Coding agent
Subcategory
Agentic AI code review and governance
Funding status
Independent, based in Boston, founded by a team of former Palo Alto Networks engineers who helped build its cloud application security business, and led by chief executive Guy Eisenkot, who earlier co founded Bridgecrew before its acquisition by Palo Alto Networks. Baz Technologies raised an eight million dollar seed co led by Battery Ventures and Boldstart Ventures, then extended it by nine million dollars to seventeen million dollars total in June 2026, with participation from AFG Partners and Disruptive VC. Its AI Code Review tool ranks first on the precision weighted Code Review Bench, and it is available on the GitHub and AWS marketplaces.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Baz integrates into the pull request workflow on GitHub and works across IDEs and the command line, connecting Figma over MCP and Jira over REST to pull design and product requirements into review. It connects to CI/CD systems, tracing, and production telemetry, is distributed through the GitHub and AWS marketplaces, and is built to coordinate other coding agents as a review and governance harness.
In practice
Your AI copilots generate diffs faster than anyone can review, and risky changes slip into production. Baz reviews every pull request with agents that understand your codebase, flag breaking changes, and enforce your standards, ranking first on a code review benchmark.
Code compiles and passes tests but silently drifts from the Figma design and Jira acceptance criteria. Baz's Spec Review agent pulls the requirements and uses browser automation to verify the delivered experience matches intent before merge.
You want to stop reviewing AI code after the fact and catch problems earlier. Baz Planner reviews the plan against your architecture before code is written, blocks unsafe paths with a risk matrix, and cuts downstream reverts and hotfixes.
Sources & related URLs
Research sources
Research notes
Added via Crunchbase agentic discovery CSV, enriched full fidelity 2026-07-07. Categorized Coding agent on AI code review and governance. Seed stage but capability dense, scoring high under straight scoring on documented multi agent architecture, memory, and integrations. Free tier plus paid plans; exact paid pricing not fully public. Independent; seventeen million dollars seed (Battery, Boldstart).
Capability coverage
11.5 / 14 capabilities · 82%
| Integrations & Tool CallingIntegrates into the GitHub pull request workflow, IDEs, and the command line, and connects Figma over MCP, Jira over REST, CI/CD, and production telemetry, Baz docs 2026-07-07 | Full |
|---|---|
| Workflow OrchestrationThe Baz platform is a central orchestration layer coordinating a multi agent review, with Baz Planner running spec, security, SRE, and fixer agents in a loop, Baz docs 2026-07-07 | Full |
| Knowledge Grounding & RAGDeeply understands the codebase through static analysis, usage patterns, and data flows, ingesting Figma and Jira specs and correlating changes with production telemetry, Baz docs 2026-07-07 | Full |
| Human Oversight & GuardrailsKeeps a developer in the loop for every big decision, suggests without blocking, and lets developers chat with the reviewer to clarify findings before merge, Baz docs 2026-07-07 | Full |
| Security, Identity & GovernanceFounded by cloud application security veterans, with an advanced security agent and a risk matrix that blocks unsafe paths and enforces boundaries before code reaches production, Baz docs 2026-07-07 | Full |
| Observability & AuditabilityIs built observable and explainable, not a black box, with clear views into stats, insights, and how the team engages with each agent's findings, Baz docs 2026-07-07 | Full |
| Memory & State PersistenceAccumulates learnings from the team automatically and modifies its own instructions, trained on past pull requests and review threads to adapt to team conventions over time, Baz docs 2026-07-07 | Full |
| Deployment & Data ResidencyDelivered as a cloud service with no on premise or self hosted deployment option documented, Baz docs 2026-07-07 | Unable to verify |
| Prebuilt Agents, Templates & PacksShips a suite of specialized prebuilt agents including Reviewer, Spec Reviewer, Advanced Security, SRE, and Fixer, configurable to each team's rules, Baz docs 2026-07-07 | Full |
| Triggers & Channel CoverageTriggers automatically on every pull request via GitHub webhooks and on every change as a pre code gateway, with manual invocation also available, Baz docs 2026-07-07 | Full |
| Model Flexibility & RoutingPowered by Amazon Bedrock foundation models with no documented customer choice or routing of the underlying model, Baz docs 2026-07-07 | Unable to verify |
| APIs, SDKs & MCP ExtensibilityExtensible through GitHub and AWS marketplaces, MCP and REST connectors, CLI, and a harness that coordinates other coding agents, Baz docs 2026-07-07 | Full |
| Testing, Debugging & OptimizationRanks first on the precision weighted Code Review Bench and validates changes in an isolated runtime, with a risk matrix evaluating every model suggestion, Baz docs 2026-07-07 | Full |
| Browser & Computer UseUses secure browser automation to click through preview environments and validate that the delivered experience matches design, though computer use is limited to validation, Baz docs 2026-07-07 | Partial |
Pricing
Free trial and on demand starter on the GitHub Marketplace; paid team and enterprise pricing not fully public
per seat team and enterprise tiers, with a free trial and on demand starter option
What is public
A free trial and an on demand starter plan are available on the GitHub Marketplace, and Baz is listed on the AWS Marketplace, but exact paid team and enterprise pricing is not fully published.
Billing mechanics
A free trial and on demand starter option, then paid team and enterprise tiers, typically per seat, available self serve through the GitHub and AWS marketplaces, with exact paid rates not fully public.
Cost watchouts
Paid team and enterprise pricing is not fully public, and cost typically scales with seats and repositories once past the free trial and starter.
Variable cost rationale
Beyond the free trial and starter, cost typically scales with seats and repositories on paid team and enterprise tiers, so it grows with team size and usage.
Additional watchouts
Start with the free trial to gauge value, then confirm per seat team and enterprise pricing directly, since exact paid rates are not fully listed.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free trial and an on demand starter plan on the GitHub Marketplace
Lowest paid plan
Starter plan with on demand reviews on the GitHub Marketplace; paid team pricing not fully public
Key ambiguities
A free trial and starter option are public via the GitHub Marketplace, but exact paid team and enterprise per seat pricing is not fully listed.
Related vendors
- Cognition — Maker of Devin, an autonomous AI software engineer
- Aider — Open-source, model-agnostic terminal coding agent that works…
- Anthropic Claude Code — Anthropic's terminal and IDE-native coding agent with sub-agent…
- Augment Code — AI coding assistant evolving into an AI-native engineering platform
- Blackbox AI — AI coding assistant for code generation and developer workflows
- Blink — Agentic AI app builder that turns a natural language prompt into a…