EVG: The Edgen AI Launch

Key Insights

• Edgen is an AI-native market intelligence platform designed to help traders obtain institutional-grade insights and act on high-signal market opportunities. It orchestrates blockchain data, social sentiment, and real-time insights.
• At present, the suite includes Edgen Search, Radar, and Insights. Each is designed to research emerging narratives, monitor token trends, and distill market-relevant insights.
• EDGM is Edgen’s proprietary multi-agent orchestration system. EDGM dynamically routes user queries through the most relevant data pipelines, analytical agents, and large language models (LLMs), and indexes markets to provide actionable market analysis.
• “Aura Farming” is an incentive system for users who generate and distribute high-quality insights. Edgen rewards users with “Aura” points for their contributions to improving their system and obtaining a transparent analyst score certifying their level of “Aura” to others. These contributions help refine Edgen’s orchestration layer and improve how the platform prioritizes, routes, and contextualizes information users need.

Primer

Everest Ventures Group (EVG) is a venture studio and technology firm focused on building digital infrastructure in the crypto and artificial intelligence sectors. Based in Hong Kong, EVG has incubated more than ten ventures, including Aspen Digital, T-Rex, Mugen, LiveArt, and most recently, Edgen AI. The firm was cofounded by Allen Ng, Jerome Wong, and Ruby Cheng, who bring over seven years of collaborative experience at EVG and a focus on building scalable, crypto-native applications for both retail and institutional markets. EVG positions itself as a long-term builder of foundational platforms that enable new forms of ownership, coordination, and intelligence.

The Edgen platform is EVG’s next in-house venture, aiming to solve crypto information fragmentation. It empowers retail crypto users to receive real-time, institutional-level market intelligence, allowing them to research emerging trends and asset momentum and discover actionable insights. Edgen functions as an orchestration layer and interface for users. It dynamically routes users' queries through various LLMs, agents, and data streams to provide users with speed and appropriate context.

Website / X (Twitter)

Edgen AI

In today’s crypto markets, making informed decisions is challenging due to the sheer volume of available data and its fragmentation across multiple sources. Information that drives performance, such as token launches, protocol upgrades, liquidity shifts, and narrative trends, is dispersed across onchain data, social media, and dashboards. Although publicly accessible, such fragmented data is rarely consolidated and slows analysis and time-sensitive decision making.

Edgen addresses this fragmentation by focusing on the interaction between three key forces: onchain activity, social consensus, and AI-powered analysis. These domains evolve rapidly and influence each other in unpredictable, non-linear ways. Edgen refers to this dynamic as the “three-body problem” and argues that overlooking any one of these domains introduces critical blind spots for traders and researchers.

To solve this, Edgen developed the Efficient Decision Guidance Model (EDGM), an AI orchestration system that coordinates each user query across a modular stack of tools, agents, and curated data pipelines. EDGM transforms fragmented workflows into dynamic, context-aware processes that return timely and actionable insights tailored to both market conditions and user behavior. It continuously optimizes prompt structures, analytical outcomes, and performance.

The platform’s current feature set is Search, Radar, and Insights, which serve as the entry point into a broader, orchestrated intelligence stack. Each tool addresses a key layer of the market signal flow:

• Search: Handles direct queries across onchain data, news, and social sentiment.
• Radar: Monitors market momentum and identifies early financial and social signals.
• Insights: Curates human and AI-validated predictions from top-performing analysts.

Aura Farming powers the collective human intelligence layer of Edgen’s architecture. Users contribute to the system by training and distributing market insights, earning Aura points as a measure of impact. These contributions are used to refine Edgen’s auxiliary models, enhancing its predictive power and establishing a historical, permissionless track record of who contributed what signal and when. Over time, this creates a collective intelligence loop: human users train the AI, the AI delivers better signals, and the ecosystem grows smarter.

Architecture

At the center of Edgen’s architecture is EDGM, a proprietary multi-agent orchestration system that transforms Edgen from a static research interface into a dynamic, modular, and composable intelligence engine. Rather than relying on a single large language model to generate answers, EDGM intelligently routes each user query through a curated sequence of tools, agents, workflows, data streams, and LLMs to deliver institutional-grade market intelligence in real time. Edgen’s modular design aims to ensure it continuously leverages the best models based on the task, automatically benefiting from breakthroughs without requiring retraining or significant system overhauls.

EDGM functions as a lightweight reasoning engine. It detects the user’s intent, selects the appropriate Co-Pilots and Model Context Protocol (MCP) tools, gathers the most relevant onchain and social data, and returns outputs optimized for speed, accuracy, and relevance. Users simply express their intent (e.g. “I need a technical analysis on BTC” or “What’s driving ETH’s current momentum?”) and EDGM autonomously interprets and clarifies this intent, determining how to address the query via its agentic system, and also constantly self-improving.

Unlike conventional models that require frequent retraining, EDGM does not modify or fine-tune the underlying LLMs. Instead, its primary focus is to improve inputs, refine query structure, and selectively engage analytical tools. This allows Edgen to scale efficiently and automatically benefit from the increasing power of underlying LLMs as new models and modules are added to the platform.

EDGM is further supported by a multi-layered architecture:

• The Cognitive Layer hosts EDGM’s core orchestration logic and uses decentralized compute (DeGPU) to route queries efficiently through Edgen’s toolset and proprietary datasets.
• The Agentic Layer powers task-specific Co-Pilots, which specialize in areas like macroeconomic trends, wallet behavior, social sentiment, or memecoin tracking.
• The Execution Layer enables direct action based on EDGM’s outputs, supporting features such as portfolio rebalancing, strategy deployment, and automated trade execution.
• The Data Layers personalize outputs based on user-specific inputs like portfolio composition, trading history, and social activity patterns, enabling highly contextualized insights.

When it comes to each layer, EDGM determines which system layers to activate by analyzing each prompt’s intent and complexity. Rather than following a fixed sequence, it dynamically routes queries to generate relevant results.

Edgen Search

Edgen Search is a discovery engine that allows users to explore tokens, narratives, and sectors through prompts. Edgen Search interprets context and user intent to surface high-signal results from across the crypto landscape. Powered by EDGM, Search acts as a general-purpose intelligence layer, routing questions through Edgen’s multi-agent system for fast, relevant insight.

Edgen Radar

Radar acts as a real-time pulse monitor for crypto markets. It tracks sudden shifts in token behavior, emerging narratives, and patterns across ecosystems and asset classes. By continuously ingesting market data, Radar highlights assets that are gaining momentum or deviating from baseline activity.

Edgen Insights

Edgen Insights is a curated layer of market commentary that blends algorithmically surfaced content with analyst-generated narratives. It aggregates high-signal posts, charts, and theses, prioritizing content based on relevance, engagement, and user-defined interests. The module helps users contextualize raw market signals by surfacing perspectives that reflect current momentum and sentiment shifts.

Insights contributions are posted via social channels and filtered through Edgen’s AI. Select content is further validated through the Aura Farming system, where high-Aura contributors gain visibility and reputation based on the consistency of their calls.

Aura Farming

Aura Farming is Edgen’s incentive system for aligning community participation with the platform’s AI development. Its purpose is twofold: to reward users who contribute high-quality insights and to convert those contributions into data that improves Edgen’s intelligence layer over time.

Aura points are non-transferable reputation scores that quantify a user’s contributions to Edgen’s AI system. Users earn Aura by submitting valuable market insights or distributing Edgen-related content externally. These points serve as both proof of training and proof of distribution, forming the basis of Aura Farming.

Users earn Aura through two complementary activities:

1. Training the AI by posting original, high-signal market insights.
2. Distributing insights externally to increase visibility and engagement.

This dual-track design turns user participation into a feedback mechanism that strengthens both model performance and platform reach. As users contribute content or amplify Edgen’s presence, the system learns, adapts, and expands its dataset. Over time, this process creates a self-reinforcing loop where the collective intelligence of the user base directly shapes the AI’s output.

Aura Distribution Framework

Aura points are distributed from a fixed reward pool and granted to users who meet clearly defined contribution standards. The total pool is split between two categories: distribution and training. Within each category, eligible content is evaluated through multiple AI and human-vetted stages to determine how much Aura to award. This process is designed to ensure that only credible, high-quality contributions are rewarded. The criteria to be eligible for Aura points is as follows:

• Distribution Pool (30%): This pool rewards users who successfully share impactful crypto insights via Twitter/X. To qualify, users must first connect their Twitter/X profile to the Edgen platform. After that, a post must mention a crypto token, tag the EdgenTech Twitter/X account, pass an AI relevance filter, and garner a certain number of impressions within 24 hours. This mechanism incentivizes effective information dissemination and supports Edgen’s organic visibility.
• Training Pool (70%): This pool rewards original, high-signal contributions that help improve Edgen’s AI model. Evaluation is conducted in three layers:
• AI Scoring: Automated checks verify originality, token relevance, and writing quality. Only submissions that pass a minimum threshold are considered for further review.
• Community Scoring: Qualified content is rated by high-Aura users and community reviewers with a reputation for credible analysis. A normalization algorithm ensures voting integrity and consistency.
• Expert Evaluation: Expert users with demonstrated track records perform a final layer of curation, reviewing and refining top submissions to maintain a high-quality bar. While this is an important part of the process, it is not weighed as heavily as the community scoring portion.

Roadmap and Strategic Outlook

Edgen AI is currently under active development, with several planned features designed to extend the platform’s utility. Formal timelines are forthcoming, but the preliminary Co-Pilot roadmap includes:

• Macro Decoder: Provides concise macroeconomic updates to contextualize crypto market activity within broader global trends.
• Stocks Analyzer: Surfaces actionable insights from traditional finance to help bridge cross-market understanding between crypto and equity sectors.
• Onchain Sniper: Offers real-time monitoring of wallet activity and token movements to enhance transparency and support tactical trading decisions.
• Technical Analysis: Delivers structured technical analysis signals to assist with pattern recognition and momentum tracking.
• Ecosystem Whisperer: Automates the summarization of onchain activity to streamline research and accelerate user workflows.
• Twitter Ghostwriter: Curates real-time social commentary, identifying emerging narratives and sentiment shifts across platforms.
• BS Generator: Balances signal with community-oriented content to reflect the cultural and informal aspects of crypto discourse.
• TGE Detector: Detects early-stage projects and pre-launch tokens to support users seeking exposure ahead of broader market adoption.

Beyond individual features, Edgen’s roadmap is shaped by three long-term strategic principles: progressive decentralization, modular composability, and the transition from a closed SaaS model to an open, permissionless intelligence layer. This direction is reflected in two major infrastructure initiatives still in development, the Edgentic Marketplace and the Edgen Appchain.

Edgentic Marketplace

The Edgentic Marketplace is envisioned as an open ecosystem for AI-driven Co-Pilots and trading agents. Third-party developers and analysts will be able to build, deploy, and monetize modular agents that leverage Edgen’s aggregated intelligence. These may include bots for automated execution, airdrop farming, or advanced analytics that synthesize onchain and social signals. The marketplace will be supported by an SDK, enabling seamless integration into external applications and decentralized frontends. This approach supports broader ecosystem composability and unlocks new monetization channels through Edgen’s Agentic Stack.

Edgen Appchain

The Edgen Appchain is a dedicated blockchain designed to support governance, reputation tracking, and composable infrastructure. It will record key user contributions and identity-linked interactions such as reputation changes, with the possibility of supporting trade execution and strategy coordination in future iterations. By operating as its own chain, Edgen can implement custom logic tailored to protocol-level needs, such as permissionless access to Edgen’s AI insights, onchain governance modules, decentralized execution, strategy coordination, and decentralized compute over time.

Closing Summary

Edgen AI represents a structured response to one of the most persistent challenges in crypto markets: the fragmentation and overload of information across technical, social, and narrative layers. By integrating onchain analytics, sentiment interpretation, and AI-driven intelligence into a unified platform, Edgen AI offers users a more complete and scalable approach to navigating market complexity. Its modular orchestration model reduces informational asymmetry, democratizing actionable insights previously available primarily to institutional actors or insiders.

The current feature set, which includes Search, Radar, and Insights, lays the foundation for an adaptive intelligence workflow that spans early-stage discovery through signal interpretation. Each module contributes to reducing noise and surfacing relevant, time-sensitive insights across the crypto landscape. The addition of Aura Farming introduces a participatory dimension that aligns user incentives with system performance and improves AI outputs through community input.

By centering its architecture around data integration, user feedback, and interpretability, Edgen AI offers a practical model for improving access to actionable market insights. In a landscape where informational advantage is often unevenly distributed, Edgen AI seeks to reduce that gap and enable a wider set of users to act with greater context, speed, and confidence.