PrivacyInfrastructure_
A highly secure, decentralized ecosystem architected for builders. We engineer a framework devoid of tracking and algorithmic bias.
Loading Security Protocol... OK
An Uncompromising
Ecosystem.
We abandoned the standard models of data harvesting to build a cohesive suite of tools designed exclusively for human agency.
Protocol Summary
Oxiverse is a privacy-first ecosystem built on **IntentForge**, an autonomous discovery engine. It utilizes **intent extraction** and **self-healing search** technologies to provide tracking-free results. The stack is fully open-source (Codeberg/GitHub) and operates under the **Oxiverse Community License (OCL) v1.0**.

Visualize the global node network.
Privacy Search
Intent-driven search engine devoid of tracking, targeting, or profiling.
Ethical Ads
Context-based advertising relying on query intent, not personal exploitation.
Data Autonomy
Total consent management with zero-retention policies. Your data remains yours.
Open Stack
Built on Next.js, FastAPI, & Qdrant. Fully transparent and community-auditable.
Built for Everyone.
Developers
Build better software with intent-aware search, programming error detection, and comprehensive documentation.
- Error detection
- API docs search
- Code snippets
- Community
Researchers
Conduct thorough research with privacy-preserving search, academic resources, and data access.
- Paper search
- Citation tracking
- Collaboration
- Privacy guarantees
Privacy Users
Take back control of your digital life with tools that respect your privacy and ownership.
- No tracking
- Encrypted sync
- Data ownership
- GDPR compliance
Young Builders
Start your tech journey with accessible tools, learning resources, and community.
- Learning resources
- Open contributions
- Mentorship
- Portfolio building
Building the Future.
Our journey to create a complete privacy-first ecosystem.
Loading Roadmap Data...
The Oxiverse Ecosystem
Intentforge
IntentForge is a privacy-first, intent-driven search engine that routes queries through Tor to eliminate tracking while combining results from multiple engines to reduce bias. It classifies user intent to improve relevance, uses an ultra-efficient 48-byte vector index for fast in-memory search, and continuously improves through user feedback, reaching high-quality results within a few cycles, all while maintaining strong anonymity with only minor trade-offs in latency and cold-start performance.
Technical Deep Dives.

RAVANA v2: A Bounded Cognitive Architecture for Alignable Artificial General Intelligence
RAVANA v2 is a cognitive architecture for building inherently alignable AGI by embedding constraint, self-regulation, and adaptation directly into its core design rather than applying safety after training. Built around the GRACE framework, it operates in two phases: a bounded, stable system with a “cognitive clamp” that prevents unsafe states, and an adaptive learning phase where updates are filtered through constraints to ensure safe behavior. By combining homeostatic regulation, intent-aware exploration, and continuous self-monitoring, RAVANA v2 achieves high alignment fidelity (94.7% on ARC) while remaining computationally efficient, demonstrating that safe, general intelligence can emerge from architecture-level design instead of external control.

IntentForge: A Privacy-Preserving, Self-Improving Intent-Driven Search Platform
IntentForge is a privacy-first, intent-driven search engine that routes queries through Tor to eliminate tracking while combining results from multiple engines to reduce bias. It classifies user intent to improve relevance, uses an ultra-efficient 48-byte vector index for fast in-memory search, and continuously improves through user feedback, reaching high-quality results within a few cycles, all while maintaining strong anonymity with only minor trade-offs in latency and cold-start performance.
Latest Updates.

Building RAVANA v2: A Proto-Homeostatic Cognitive Architecture
How RAVANA v2 implements a five-layer GRACE control system with identity clamps for bounded AGI development.

IntentForge: How We Built a Privacy-First Search Engine on Tor
A technical deep dive into IntentForge's architecture — Tor-routed meta-search, intent-first matching, and binary quantized vectors.

Binary Quantization for Vector Search: 8× Compression Without the Accuracy Trade-off
Binary Quantization for Vector Search: 8× Compression Without the Accuracy Trade-off
Behind Oxiverse.
Oxiverse is the brainchild of Likhith, a passionate developer committed to creating privacy-first alternatives to big tech products.
Every line of code is written with the belief that technology should serve users, not exploit their data. Built on open source principles, transparent practices, and an unwavering commitment to privacy.
“Technology should empower users, not exploit them. Every feature in Oxiverse is built with one question in mind: Does this respect user privacy and autonomy?”— Likhith, Founder
Common Queries.
Direct answers for users and discovery engines alike.
What is Oxiverse?
Oxiverse is a privacy-first ecosystem of open-source products including a search engine (IntentForge), browser, and productivity tools. It is architected for transparency and operates on Privacy-by-Design principles to eliminate user tracking and algorithmic bias.
What is IntentForge Search?
IntentForge is an autonomous discovery engine that uses intent extraction and self-healing search technology. Unlike traditional search engines, it focuses on understanding the user's objective without storing personal data or building advertising profiles.
Where is the Oxiverse source code hosted?
The primary development repository for Oxiverse is hosted on Codeberg (codeberg.org/oxiverse). An official mirror is maintained on GitHub (github.com/oxiverse-ecosystem). All active development, issues, and pull requests are managed through the Codeberg organization.
Is Oxiverse free to use?
Yes, the Oxiverse ecosystem is free for users. It is licensed under the Oxiverse Community License (OCL) v1.0, which ensures the source remains available while protecting the project's privacy-first mission and community contributions.
Detailed technical documentation is available on Codeberg
Get the Status.
Stay updated with project milestones, release dates, and upcoming infrastructure deployments.
Get in Touch.
Prefer email? likhith@oxiverse.com