Why Angelo and I Brought Vice Alliance Back After 18 Years

In 2008, Angelo Manzano and I started a studio called Vice Alliance. Two people. One contract. No producer, no account layer, no junior strategist who didn’t read the brief. We worked out of Miami and shipped for C-SPAN, ABC News, Good Morning America, WWE, Seagate, The Daily Business Review, and a list of smaller clients who, in aggregate, paid for our kids’ shoes.

We never officially closed it. The work just stopped having that name on it. I went to Tribune Digital as a Software Engineer III and led the AWS migration for six East Coast newspapers. Angelo kept designing for enterprise clients. We stayed friends. We stayed each other’s first call when something on the work was actually hard. We just didn’t put it under a brand.

Eighteen years later, in 2026, we restarted it. This is the honest story of why.

The agency model is broken in a specific way

Anyone who has been on the buyer side of a $250K agency engagement has watched the same thing happen. The pitch is two senior people in a room. The contract gets signed. The senior people disappear into a Slack channel. The work gets handed to a junior strategist, an account manager, two designers you’ve never met, an offshore dev shop, and a producer whose job is to put status meetings on your calendar.

The strategy deck gets approved, then handed to people who weren’t in the room when it was decided. It gets translated into Figma frames that don’t reflect what was actually agreed to. Those frames get handed to engineers who simplify whatever they can’t ship in a sprint. By the time the site is live, the work has been through eight pairs of hands, and not one of them has read the original brief in three months.

Clients I respect describe this the same way every time: what we got was nothing like what we were sold.

That’s not because agency people are bad at their jobs. The model can’t preserve intent across that many handoffs. Every layer between the strategy and the production code is a place where the work loses information.

Why two principals beat twelve

The pitch on the rebuilt vicealliance.com is one sentence: strategy that survives shipping. The promise underneath is shorter: two principals, one contract, we don’t split design from engineering.

That’s not a marketing line. It’s the operating model. Angelo and I are both in the Figma file, both in the Astro repo, both on the call with the client. When a strategic decision needs to become a frontend implementation, the strategist and the engineer are the same two people they were in week one.

No producer to coordinate us, because we’re two people. No account manager to translate, because we don’t need translation. No senior reviewer to catch what the juniors missed, because there are no juniors.

The trade-off is obvious. We can hold a small number of programs at a time, most starting near $25K. We can’t take everything. We take what we can ship well.

The trade-off is also the feature. What you agree to is exactly what ships.

What an Agentic Pool Made Possible That Wasn’t Possible in 2008

In 2008, a two-person studio could ship a brochure site or a microsite. We could not, with any sanity, ship a full enterprise platform. The math didn’t work. You needed a team.

In 2026, the math works.

Here’s why. The cost of execution, actually producing the code, the typography, the responsive states, the schema markup, the SEO infrastructure, the accessibility audit, the deployment pipeline, has collapsed by something like an order of magnitude. Not because the work got easier, but because the right pool of agents, coordinated properly, lets two senior people ship what used to require a team of ten.

We’re not running one tool. We’re running a working pool of frontier agents. Claude handles long-form reasoning, code architecture, and the slow careful work of design systems. Gemini handles multimodal review and live-data integration. Codex carries the raw production-code volume. Higgsfield handles visual generation and brand imagery. A rotating bench of specialized models covers transcription, embeddings, schema validation, and SEO scoring. None of these would do the job by themselves. Together, they do most of the keystrokes.

The piece I’ve been building OpenClaw for the last year is the harness that orchestrates that pool. It’s not a chatbot. It’s not a prompt library. It’s a system that lets Angelo and me move through design, copy, code, schema, and deployment without losing the thread between any of them, and without us having to remember which agent is best at which subtask. The strategy from the kickoff call ends up in the frontmatter of the markdown files. The brand tokens from the design system end up in the Tailwind config. The schema graph from the SEO plan ends up in the production HTML on every page. The right model handles the right step, and the system holds the line so we don’t have to.

The result: a server-rendered Astro production site that used to take twelve people three months now takes two people three weeks. Same quality, same SEO, same accessibility, same performance budget. Less drift, because there are fewer hands, and the hands that remain are augmented by an orchestrated agent pool rather than replaced by a single brittle one.

The 2008 version of Vice Alliance couldn’t have done this. The 2026 version can.

What hasn’t changed

The thing I keep telling people about Angelo is that he’s a senior UX designer who actually understands engineering constraints. He doesn’t ship me a Figma file that requires me to invent two new framework features to implement it. I don’t ship him an engineering decision that quietly torpedoes the brand system.

We’ve been calibrating to each other since 2008. That doesn’t unlearn. The first project we shipped under the relaunched brand felt exactly like the projects we shipped fifteen years ago: fast, calm, almost boring in how little drama there was. We didn’t need to relearn how to work together.

The tooling changed. The partnership didn’t.

Who this is for

If your last digital build went sideways (the kind of sideways where the agency cashed your check and shipped you something that doesn’t represent what you bought), Vice Alliance is probably the right kind of overcorrection.

We work best with clients who:

• Have a real budget. Floor is around $10K. Most programs are $25K and up. We don’t do speculative work.
• Want senior people on the work. If you want a team of fifteen on a status meeting, we’re not the right studio.
• Are okay with a small number of programs at a time. We don’t take everything. We take what we can ship well.
• Came in through an introduction or direct outreach. We don’t run cold inbound. The model only works at the volume we can hold.

If that’s you, the email is hello@vicealliance.com. There’s no form gauntlet, but that doesn’t mean your message lands in a dead inbox. The email is a gateway into a 100% agentic-assisted project management system wired into Discord, Telegram, and email. Your first message gets triaged, routed, and turned into a working project channel where Angelo, me, and the relevant agents in our pool can collaborate with you in whichever surface you actually live in. You don’t get a chatbot. You don’t get a 30-minute discovery slot on a calendar. You get a project channel with two principals in it, and the agent pool quietly working in the background.

What we’re shipping

The relaunched studio site features four cases by name. Beat Saber (VR/gaming UX). Marriott Vacation Club (enterprise hospitality). FishIntel.ai, the AI fishing intelligence SaaS I built from concept to live product in 24 hours. And TheGlobe.com (NASDAQ: TGLO), the foundational web work that produced the 606% first-day IPO record and two co-invented U.S. patents (US 6,999,458 internet telephony and EP 1836591 A2 web-based instant messenger). The full archive is eighteen cases.

One long-term retainer, Robbie’s of Islamorada running since 2016, has generated more than $7.1M in online revenue under the partnership. That’s the operate-retainer model in action: project becomes program, program becomes infrastructure, infrastructure compounds.

The full project breakdown lives in my portfolio on this site.

The honest pitch

I’ll close with the thing I told Angelo on the phone the day we decided to do this.

There are a lot of studios that say they ship great work. There are very few studios where you can name both people who are going to be on your project before you sign the contract, and trust that those two people will still be on your project when it goes live. We’re one of the few. That’s not a marketing claim. That’s what happens when there are two of you and zero account layer.

We were already doing this work without the brand. Putting the brand back on it is a way of being honest about how the work actually happens, and a way of telling new clients what they’re going to get.

Strategy that survives shipping. That’s the whole pitch.

If it sounds like what you’ve been looking for, the door is at vicealliance.com. Miami. Florida Keys. Orlando. Melbourne. Wherever the work needs to happen.

Welcome back, Angelo. Let’s go.