Skip to content

Roadmap

v1.0 is built. Here is what comes next.

The committed near-term platform is complete and merged. The long-term vision has four tiers; this is where each one stands.

  1. v1.0Shipped

    Repository Intelligence Platform

    The committed near-term platform, delivered as three tracks and each closed against its own definition of done.

    • Core platform: 12-language parsing, four storage backends plus in-memory, retrieval, memory, workflows and agents
    • Compression & cost: the compression engine, Compress-Cache-Retrieve, the OpenAI/Anthropic proxy, savings tracking
    • Graph intelligence: confidence tags, the Neo4j graph engine, reflection and learning, full REST, MCP and CLI parity
    • The web console for people, served by the API
  2. v2.0Planned

    Enterprise AI Platform

    Running BBM-Atlas for many teams at once. Written down, deliberately deferred past the MVP.

    • Multi-tenancy
    • Single sign-on
    • Role-based access control
    • High availability
  3. v3.0Planned

    Distributed Engineering Intelligence

    Reasoning that crosses repository boundaries.

    • Cross-repository reasoning
    • Distributed agent clusters
  4. v4.0Vision

    Engineering Operating System

    The long-term ambition, with no committed scope yet.

    • Autonomous refactoring
    • Self-healing workflows

Alongside v1.0

Hardening that continues

None of this blocks using the platform: it is scale, operations and deliberately deferred scope.

  • Performance at production scale

    Calibrating against real Neo4j graphs at the 100k-node target; impact analysis still misses its latency target at that ceiling.

  • Failure-log scanners per agent

    Real scanners for Claude Code, Codex, Gemini and OpenCode logs, waiting on real sample data.

  • Production deployment

    The defaults are safe, but TLS, a secrets manager and orchestration are still to come; rate limiting is per process.

  • Deferred compression scope

    Embedding-based relevance scoring, ML content detection and OpenTelemetry/Langfuse telemetry.

How it is built

Audited, tested, and tested again

Every change lands through a pull request with regression tests, and CI runs the full suite plus real Postgres, Neo4j, Qdrant and Redis backends and a Docker smoke test.

merged pull requests
104
automated tests
~2,800
line and branch coverage
93%
real backends tested in CI
4