Build a high-conviction position in Akash Network (AKT), as it serves as the primary infrastructure play for decentralized AI compute and is currently a top portfolio priority. For high-risk "VC-style" upside, allocate smaller amounts to Avvo (AVVO) on the Solana ecosystem to capture the growing AI agent narrative. Consider a speculative micro-cap position in Inferix (INFRA) while it sits at a low market cap, but treat it as a high-volatility bet that requires extreme patience. To manage risk effectively, plan to exit your initial capital after a 2x or 3x gain and look to fully realize profits within a 9 to 18-month timeframe. Focus your investment on the compute and distribution layers of AI rather than individual models, as hardware access is becoming the more valuable commodity.

Investment Thesis: The AI Super Bubble

The speaker believes we are entering an AI Super Bubble over the next couple of years. The core strategy is to build a crypto AI portfolio now, during market "fear" and the perceived bottom, before the general public drives prices significantly higher.

Key Strategy Takeaways

• Risk Management: The speaker treats micro-cap tokens as "VC-style bets," expecting 8 out of 10 to go to zero. Success depends on identifying the 2 out of 10 that provide "alpha" (high returns).
• Profit Taking: A major lesson learned from previous cycles is the necessity of selling. The speaker aims to take initial capital out after a 2x or 3x gain and take profits aggressively along the way.
• Time Horizon: On average, the speaker’s successful picks peaked 9 months after the initial investment, with some taking up to 18 months.
• Current Stance: "Nibbling" at positions slowly rather than rushing in, anticipating a few months of market settling.

Akash Network (AKT)

This is the speaker's largest single position in his entire crypto portfolio. He recently sold his Bitcoin (BTC) and Ethereum (ETH) to increase his stake in AKT.

• Core Function: A decentralized, permissionless cloud compute marketplace (often described as "decentralized AWS").
• Longevity: The founder has been building the project since 2015, providing a level of "institutional-grade" maturity that newer projects lack.
• The "Inference" Opportunity: While high-end GPUs (H100s) are needed for training AI, AKT can utilize consumer-grade GPUs (like gaming PCs) for inference (running the AI models).
• Censorship Resistance: As governments potentially clamp down on AI model access (e.g., Anthropic/OpenAI), decentralized compute becomes a "credible fallback" that cannot be turned off.

Takeaways

• Bullish Sentiment: Extremely high conviction; viewed as a fundamental infrastructure play for the AI era.
• Market Cap: Mentioned at approximately $220 million with a favorable Fully Diluted Valuation (FDV) due to its age.
• Key Use Case: Targeting enterprise users who want to run open-source models without sending data to centralized providers like Anthropic or Google.

Avvo (AVVO)

A micro-cap "VC-style" bet that the speaker has held through the bear market.

• Context: Built on Solana (SOL), which the speaker identifies as the leading chain for capturing market attention.
• Focus: AI agents that perform trading tasks and utilize tools like Claude and Codex to automate investment strategies.

Takeaways

• Bullish Sentiment: High, specifically due to the "AI Agent" narrative which is currently popular in crypto circles.
• Action: Holding as a high-risk, high-reward play within the Solana ecosystem.

Inferix (INFRA)

Another micro-cap token held in the "VC bucket" that has remained in a downtrend throughout the bear market.

• Context: A very small project with a market cap around $2 million.
• Development: The speaker maintains direct contact with the developers and notes they have continued building despite poor price action.

Takeaways

• Sentiment: Speculative but high conviction in the team's talent.
• Risk Factor: Extreme volatility and low liquidity; it is a "bet" on a recovery when market momentum returns to AI.

Investment Themes & Honorable Mentions

Decentralized Inference

• Project Mentioned: Nightfall (specifically a project discovered by community members).
• Context: A project that achieved a 43x return (from $300k to $13M market cap) by optimizing decentralized inference across scattered GPUs.
• Insight: Even in a bear market, the "speed" of decentralized AI is improving, catching the attention of major players like NVIDIA.

Non-AI Holdings

• The speaker confirmed he still holds ResearchCoin (RSC) and Geodnet (GEO), though they were not the focus of this AI-specific update.

Risks Mentioned

• Liquidity Risk: Some micro-caps may show massive percentage gains on paper (e.g., 500x), but lack the liquidity to actually sell and realize those profits.
• Model vs. Compute: The speaker suggests value may shift from the AI models (which are becoming open-source and commoditized) to the compute and distribution layers (like Akash).