Andreessen Horowitz is doubling down on the unglamorous work of getting AI to actually function inside real businesses. The venture capital giant just appointed Andrew Omori, its Partner and Head of Fund Strategy, to the advisory board of AZIO AI, an artificial intelligence infrastructure company preparing to merge with Envirotech Vehicles (NASDAQ: EVTV). The move reveals where a16z believes the real money in AI will be made over the next few years: not in building the next ChatGPT, but in the messy, complex work of deploying AI systems that actually solve problems for enterprises.

Omori brings two decades of capital markets and M&A experience, including work on major tech exits like Coinbase and OpenAI, as well as IPOs for Snap, Alteryx, and GoDaddy. At a16z, he leads strategic initiatives around capital allocation, financial modeling, and fund operations. His appointment signals that AZIO AI is preparing for serious institutional-scale operations as a public company focused on GPU distribution, AI data center capacity, and compute infrastructure for next-generation AI workloads.

Why Enterprise AI Deployment Is Becoming the Real Battleground?

The appointment comes as a16z is simultaneously launching the FDE Fellowship, an eight-week cohort for Forward-Deployed Engineers and Applied AI leaders at the forefront of deploying AI into real-world enterprises. The fellowship, which begins in July 2026, brings together practitioners from companies like Decagon, ElevenLabs, Cursor, Harvey, Google, Snowflake, Hex, Rippling, and OpenAI to share what's actually working in the field.

The FDE role originated at Palantir in the early 2010s with a simple insight: building great enterprise software requires deep customer alignment. Today, that role has become critical for agentic platforms, which are AI systems designed to take actions autonomously within a business context. Unlike generic AI chatbots, these agents must be woven through complex webs of tools, automations, and integrations specific to each industry and customer.

"FDEs are at the frontlines, engaging with the customer to discover, integrate, and build under tight timelines and with emerging expectations for highly-personalized software," a16z explained in announcing the fellowship.

a16z, Introducing the a16z FDE Fellowship

The best FDEs combine technical depth with customer empathy to navigate complex enterprise organizations, earn stakeholders' trust, and deliver measurable business outcomes. For AI-native products, the work happening in the field right now isn't just deployment; it's research and development. FDEs are discovering what works, what doesn't, and how to make AI agents actually useful in the messy reality of enterprise workflows.

How to Identify Where AI Startups Can Actually Win?

A16z partner recently outlined a framework for understanding where AI application companies can survive and thrive. The key insight: avoid competing directly with OpenAI and Anthropic on their turf. Instead, build in what a16z calls "the rest of Oz," referring to complex, vertical-specific problems that horizontal AI platforms can't easily solve.

• The Yellow Brick Road (Dangerous Territory): This is where OpenAI and Anthropic are walking, building generic AI coworkers with access to standard tools like Google Drive, Slack, Salesforce, and Notion. Startups attempting this playbook with the same connectors and no distribution face an uphill battle against labs with massive brand halo and architectural advantages.
• The Rest of Oz (Opportunity Zone): These are complex, multi-step, multi-player workflows that require deep industry knowledge, legacy system integration, and deterministic outcomes where ambiguity isn't acceptable. Think legal redlines, insurance underwriting, or sales development representative campaigns that need human approval at multiple stages.
• Defensibility Through Accumulated Knowledge: Application companies can build competitive moats by internalizing unwritten industry norms, undocumented standards, and tribal knowledge that lives in practitioners' heads but never appears in public training data. A company that has run agents through a hundred legal redlines or a thousand insurance underwriting cycles has internalized the shape of the problem in ways new entrants cannot replicate.

The labs themselves are acknowledging this reality. OpenAI and Anthropic have announced massive forward-deployed joint ventures to build entire companies around configuring and customizing their models for enterprise customers. As a16z noted, you don't pour billions into those programs if you think the next model release will solve everything.

Omori's appointment to AZIO AI's advisory board reflects this strategic shift. AZIO AI is focused on the infrastructure layer, providing the GPUs, servers, and data center capacity that these enterprise AI deployments require. As the combined company prepares for public markets, it will need the kind of capital markets discipline and M&A experience Omori brings. The company is scaling GPU and server deployment, including NVIDIA B200 hardware deliveries, building modular AI data center capacity with behind-the-meter power, and converting a growing customer pipeline that has already received deposits.

The broader message from a16z is clear: the next wave of AI wealth creation won't come from building better chatbots. It will come from the unglamorous, customer-intimate work of making AI systems actually function inside the enterprises that need them most. That's where the FDE fellowship is focused, and that's why a16z's top fund strategist is now advising an infrastructure company preparing to go public.