AI is a Great Producer. It's a Terrible Creative.

Krish Menon has built brands for Equinox, PepsiCo, and some of the most recognized companies in the world. He is not anti-AI. But he will tell you, without hesitation, that humans are still absolutely necessary.

"AI still sounds like AI," he said on a recent episode of Building for Others. "AI still feels like AI."

Controlling for that has shaped how Menon built two internal tools for his brand consultancy, Angry Gods, that are completely changing how his team works.

The chaff and the gold

Menon's framework is simple. Every creative engagement has two layers: the gold, which is the insight, the idea, the creative leap that only a sharp human mind can make, and the chaff, which is everything else: research, competitive analysis, framework construction—the hours of groundwork that have to happen before anyone can think clearly.

"The chaff for us is all of the grunt work that goes into researching and understanding and pulling together constructs for a company," Menon explained. "There are things you can frameworkize, and that's mostly strategy work."

His argument is that AI is exceptionally good at the chaff. It can synthesize research, surface competitive patterns, and structure strategic frameworks faster than any team of analysts. What it cannot do is connect the dots that nobody expected to connect. It cannot do weird thinking.

"AI is not meant to do weird thinking," Menon said. "It does weird thinking, but based off of what it thinks is weird. There's no craziness inside of AI because it's not been trained to do that."

The implication for agencies, and for any knowledge-work organization, is that the question isn't whether to where to aim AI, not whether or not to use it. 

Ghost: a brand strategy command center

To put that philosophy into practice on client work, Menon built Ghost, an internal brand strategy tool that functions as what he calls "an actual copilot" for brand strategists.

Ghost is a structured dialogue system built around 22 of the most widely used brand strategy frameworks, from Four C's Analysis and Brand Pyramids to Simon Sinek's Golden Circle and StoryBrand's Seven-element Brand Ccript. Rather than asking a strategist to stare at a blank page, Ghost opens a conversation, it asks questions, it pushes back, and it builds strategic artifacts.

"I wanted to have someone sitting side by side with you as a strategist," Menon said. "A smart thinking partner that would ask the right questions, go out and find the right information, pull the data together, but let the strategist make conscious decisions."

The result is a system where the AI does the heavy lifting on research and structure, while the human retains authorship of the actual strategic thinking. Ghost can run a full brand strategy engagement, a repositioning, a messaging framework, a naming exercise, or a brand audit, and it can convert outputs between frameworks when a client comes in with a preference for a specific methodology.

Critically, Ghost is designed to disagree. It questions inputs before building on them and offers variations. It stress-tests coherence across different strategic elements and flags where things don't hold together. That's a deliberate design choice, and one that separates it from AI tools that simply validate whatever the user puts in.

Athena: a talent network built for the freelance era

The second tool Menon built solves a different problem, one that sits upstream of any client engagement: finding the right people.

Angry Gods operates with 65 collaborators and just two full-time employees. That model, which Menon developed deliberately after selling his previous 250-person agency, Phenomenon, depends entirely on being able to identify and mobilize elite freelance talent quickly. The existing infrastructure for that—job boards, LinkedIn posts, Behance searches—simply wasn't built for it.

"Every good freelancer is always taken," Menon said. "Figuring out when they have an opening is hard because there's no central repository for this. There is no LinkedIn for agency freelancers."

Athena is his answer. It is a freelancer management platform that combines an applicant-facing profile system with an AI-powered talent scout. Freelancers apply, build profiles, list their skills, rates, and availability, and are assessed against Angry Gods' internal criteria. The platform then surfaces candidates for specific briefs, explaining not just who matches but why, and where there might be gaps.

The scout function searches across LinkedIn, Medium, Substack, and other sources in real time, matching candidates to open briefs and ranking them by fit. In the month before the podcast recorded, Angry Gods had run around 15 searches through Athena and hired seven of the people who appeared in the top three results. In one case, a friend agency ran a search for a role they had been trying to fill for six weeks. Athena surfaced the person they eventually hired in under two minutes.

Athena also handles newsletters to keep the freelancer network warm, internal notes and ratings on candidates, and the full contract workflow once someone is brought on. 

Menon is weighing whether to release Athena as a SaaS product. Two agency CEOs offered him $100,000 a year for access after seeing a demo. The hesitation is strategic: if Athena becomes the industry standard for finding elite freelance talent, it erodes one of Angry Gods' core competitive advantages.

Freelancers can apply to join the Athena network at athena.angrygods.com. Agencies interested in learning more can reach Krish directly at kmenon@angrygods.com.

The broader point

What Menon has built at Angry Gods is a working model of something a lot of organizations are still theorizing about: AI that handles the process so humans can focus on the judgment.

The tools are different in scope, one faces inward toward talent, the other toward client work, but they share the same underlying belief. AI is most valuable not when it replaces human thinking, but when it clears enough of the path that human thinking can actually happen.

That belief is increasingly hard to argue with. The companies seeing real returns from AI are not the ones that handed their teams a chat interface and hoped for the best. They are the ones that figured out where the chaff was, aimed AI at it, and freed their people to do the work that actually required them.

Sachin and the team at elvex have been making the same argument for enterprise organizations at scale. elvex is the proactive enterprise AI platform built to bring that kind of leverage to every knowledge worker, not just the power users who would have figured it out anyway. Learn more about elvex.