AI platforms and open-source projects.
What Ambharii Labs builds, and which parts can be adapted to your environment.
Products
Platforms and open-source projects built at Ambharii Labs. Several can be customised and deployed to a client’s requirements as part of a consulting engagement.
GenomixIQ
A multi-tenant agentic AI platform for genetic testing, clinical diagnostics, pharmacogenomics and precision medicine.
Product page →ARIA RCM
AI agents for healthcare revenue cycle management, covering denial prediction and accounts-receivable workflows.
Product page →AgentMesh
An AI governance proxy that sits in front of AI tools. Open source (Apache-2.0), and available to be customised and deployed to a client’s needs.
Product page →Bulwark
A Python framework with a five-layer defence for AI agents against prompt injection. Apache-2.0. Designed to support audit-trail and access-control requirements; it does not certify compliance.
Product page →Keelgate
A safety-first agent harness for AI agents that touch money and regulated actions. The model proposes; deterministic code decides, using capability-scoped tools, policy-as-code gates, human approvals and a tamper-evident audit trail. Apache-2.0.
GitHub →Tycheon
A calibrated forecasting and risk layer for financial time series: conformal prediction intervals, forecast-to-risk metrics and leakage-proof backtesting. Alpha, Apache-2.0. For research and risk analytics, not investment advice.
GitHub →ARGUS-AI
Python library (PyPI: argus-llm) that computes G-ARVIS scores for LLM outputs with rule-based heuristics. Beta, Apache-2.0.
Product page →Need it tailored?
Product pages describe what each project is today. If you want one adapted to your environment, data, or workflow, that is a consulting engagement: we scope it together, then I build and hand it over.
Start a conversation
Tell me which product you are interested in and what you need it to do.
Contact Ambharii Labs →