Skip to content

Evaluate one OSS or self-hosted provider alternative #2

Description

@kmishra1204

This product foundation intentionally keeps paid-provider coupling visible instead of hiding it.

For this repo, the current paid or hosted risk areas include:

LLM APIs, hosted realtime backends, hosted auth, hosted deployment

Candidate OSS/self-hosted directions already identified:

OpenHands/SWE-agent style runner evaluation, PostgreSQL event store, Docker or Firecracker-style sandbox direction, Ollama/vLLM adapters

What would help

Pick one provider risk and produce a focused evaluation:

  • what OSS or self-hosted option looks credible
  • what integration boundary would need to change
  • what security or data-retention impact exists
  • what local proof would be enough before implementation

Good outcome

A short markdown note, small adapter spike, or issue comment with verified evidence is valuable. The goal is practical contributor learning, not a broad rewrite.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enterprise-hardeningSecurity, reliability, observability, governance, or production readinesshelp wantedMaintainer-approved help is welcome.oss-alternativeOpen-source or self-hosted replacement research or implementation

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions