AP/Parate®

Work

Three rooms.
One obsession.

Each role taught me a different thing about how distribution actually works. Here's what I built and what I took away.

01 / Dec 2025 - Present

Kasparro

GTM Sales, Content & Community Lead

Owning the GTM motion across content, outbound, community, founder branding, and sales enablement.

  • Built long-term processes for content, outbound, sales enablement, and reporting that streamlined how the team operates and cut execution time across the board.
  • Scaled two AI persona accounts on X from zero to 400+ followers and 600K+ impressions each in 30 days.
  • Grew Reddit accounts to 18K+ and 16K+ karma, pulling inbound interest from SEO and marketing communities.
  • Run 5,000+ leads a month through outbound: Apollo, Clay, Instantly, HeyReach, with personalization at every step.
  • Built relationships with CMOs at D2C brands that started as cold emails and turned into real pipeline.
  • Produced brand audits, sales decks, and GTM recommendations that improved client conversion.

What I learned

"Distribution is the actual product problem. Everything else compounds once you get that right."

02 / Aug 2025 - Oct 2025

Creator Chart

Founder's Office Associate

Three months wearing every hat.

  • Ran SEO with Ubersuggest and built out AEO and GEO experiments to surface the brand inside ChatGPT, Perplexity, Grok, and Claude, work most marketing teams had not started yet.
  • Executed full GTM motion through webinars and LinkedIn outreach, identified the ICP, and closed a ₹1.5L client.
  • Built content strategy across Instagram, LinkedIn, YouTube, Reddit, Medium, and Substack to drive consistent growth.

What I learned

"AI search is the next decade of discovery. The brands optimizing for it now will own it later."

03 / Apr 2025 - Aug 2025

Unacademy

Growth and Content Strategy Operations Associate (YouTube)

Grew Unacademy's official GATE YouTube channel by 30% through content optimization, performance tracking, and engagement work.

  • Pushed the channel into the top 10 list of YouTube channels in India for the GATE category.
  • Used YouTube as a primary sales funnel and lifted high-intent leads by 34%.
  • Ran deep competitor analysis to find discoverability gaps and outperform players in the same domain.

What I learned

"You don't need more content. You need better positioning."

Toolkit

What I reach for when the work starts.

Programming

PythonClaude Code

Growth & outbound

ApolloClayInstantlyHeyReachMailchimpCold emailLead generationCampaign optimization

Automation

N8NExcel

Search & research

SEOAEOGEOUbersuggestCompetitor research

Content & distribution

LinkedInXInstagramRedditYouTubeMediumSubstack

Core

Growth thinkingProblem-solvingCritical thinkingRapid experimentationProcess building

In action

See work in action.

I don't just talk about AI persona accounts, I build and grow them. Three accounts, built from zero, each with its own voice, audience, and growth curve. Here's the proof.

AI persona accounts, X analytics (3M view)

X Analytics for The Agentic Co.

The Agentic Co.

@AICommerceGuy_

3 months

84.5K

Impressions

78

Followers

5.3%

Engagement

Niche AI commerce account. Small audience, high engagement: 5.3% rate with 2.6K likes and 359 replies. Proof that a tight niche beats vanity metrics every time.

X Analytics for Aisha Gupta

Aisha Gupta

@Aishaguptaa_

3 months

1.3M

Impressions

482

Followers

3%

Engagement

From zero to 1.3M impressions in 3 months. 29.4K likes, 528 reposts, 904 replies. The growth spike in April wasn't luck, it was a system kicking in.

X Analytics for lina chen

lina chen

@linachen_nyc

3 months

944.8K

Impressions

540

Followers

2.7%

Engagement

Nearly 1M impressions, 540 followers, 292 reposts. Shares up 966%. Built the same playbook, different persona, and it worked again.

All three accounts built and operated by me at Kasparro. No paid promotion. No follow-for-follow. Just content, timing, and systems.

Writing

Thread12 tweets

The agentic commerce timeline, everything I've watched unfold since 2022

From academic papers to $385B projections. A thread tracking every inflection point in AI-powered commerce: Perplexity's quiet launch, OpenAI's Instant Checkout, Amazon's panic, and why your product data is the only variable left.

1

I've been watching the agentic commerce space since 2022. Back then it was academic papers and prototypes. Nobody was paying attention. Three years later, AI agents are completing real purchases inside chat windows. $385 billion projected by 2030 in the US alone. Here's everything I've watched unfold and where I think this goes next.

2

2022 to 2023 was quiet. ChatGPT launched and researchers started asking interesting questions. Can LLMs influence what people buy? Can an AI shopper mimic real buyer behavior? Amazon was publishing papers on grounding agent responses with citations. I remember reading these and thinking: this changes commerce if anyone actually builds it. Nobody had. Yet.

3

November 2024 was the moment most people missed. Perplexity shipped Buy with Pro. Real purchases. Inside a chat. PayPal handling the money. Over 5,000 merchants connected. No fanfare. No headlines. But that was the first time a customer bought a product without ever opening a browser, visiting a store, or clicking an ad. I knew right then the funnel was broken.

4

September 2025 is when it got loud. OpenAI and Stripe launched ACP with Instant Checkout inside ChatGPT. Etsy, Walmart, Target, Sephora, Best Buy, all connected. 800 million weekly users suddenly had a buy button in their chat. Etsy stock jumped 16% in a day. That number told me the market understood this wasn't experimental anymore.

5

October 2025. McKinsey put a price tag on it. $3 to $5 trillion globally by 2030. Up to $1 trillion in US retail. Called it comparable to the web and mobile revolutions. Except faster because AI agents ride the same rails humans already built. Morgan Stanley said $385 billion. Bain said $500 billion. When three firms independently converge on the same conclusion, it's not a prediction. It's a trajectory.

6

November 2025. Amazon panicked. Sued Perplexity for unauthorized agent purchases. Blocked OpenAI's crawlers. Removed 600 million products from ChatGPT's reach. Updated legal terms to ban agent behavior entirely. I watched this and realized: Amazon just pushed every AI shopping agent toward independent brands. If you're a DTC merchant with clean data, Amazon's wall just became your advantage.

7

January 2026. NRF. Everything accelerated. Google and Shopify launched UCP with Walmart, Target, and 20+ partners. Microsoft launched Copilot Checkout on ACP. Shopify announced Agentic Storefronts. One toggle in your admin to sell across ChatGPT, Google, Copilot, and Perplexity. Two protocols. Four platforms. One month. I've never seen commerce infrastructure move this fast.

8

March 2026. First reality check. OpenAI pulled back Instant Checkout. Performance wasn't there. Pivoted to ChatGPT Apps where merchants control their own checkout inside the chat. But Google doubled down. Expanded UCP with multi-item carts, real-time catalog queries, and loyalty integration. Shopify activated Agentic Storefronts for all US merchants. 5.6 million stores connected overnight.

9

April 2026. The payment layer landed. Visa launched Intelligent Commerce Connect. Amex launched ACE with Agent Purchase Protection. Mastercard launched Agent Pay. Alibaba integrated Alipay into Qwen for 300 million users. Five separate payment systems for AI agents. All within months of each other. When every major payment network builds agent infrastructure simultaneously, that's not experimentation. That's a bet.

10

Where we sit right now. May 2026. 900 million weekly ChatGPT users. 75 million daily Google AI Mode users. AI referral traffic to retail up 4,700% year over year. 23% of Americans have bought something through AI in the past month. 68% of consumers used AI tools while shopping in the last three months. But only 13% have completed a full end-to-end agent purchase. That gap between comfort and action is the entire opportunity.

11

Here's my prediction. We're at the end of the infrastructure phase. Protocols exist. Payment rails exist. Checkout surfaces exist. The plumbing is done. 2027 will be the consumer awareness tipping point. AI agent purchases will go from 13% to 30%+ adoption. By 2028, Gartner says 20% of all ecommerce transactions will be executed by AI agents. By 2030, this is a $385 billion market in the US at minimum. $5 trillion globally at the high end.

12

The pattern I keep seeing from four years of watching this space: Every brand that prepared early for a platform shift captured disproportionate share. Every brand that waited played catch-up at a premium. Agentic commerce is following the same curve. Except compressed. What took mobile 10 years will take agents 3 to 4. The infrastructure is built. The protocols are live. The payment rails are connected. The only variable left is your product data.

Analysis

Nobody is talking about the chargeback problem in agentic commerce

AI agents are making purchases on behalf of customers. When the agent gets it wrong, wrong size, misread policy, unwanted item, someone has to pay. Right now the answer is: the merchant.

Infrastructure

Your CMS was built for humans. AI agents need something completely different.

Structured API endpoints. Real-time inventory feeds. Machine-readable product attributes. Legacy platforms like Magento, Miva, and NetSuite weren't designed for any of this.

Videos

Humans aren't your only customers anymore. Your systems need to be built for agents, that's the real shift happening right now.

Most people use AI the wrong way. The only thing that matters is taste and judgement, knowing what to double down on.

YC startups doing thirst traps and using that for distribution. Yes, really. And it's working.