What I build

  • Email triage agents that sort, label and draft replies for a person to approve.
  • Document processing: an agent reads contracts, forms or report cards and fills your spreadsheet or CRM.
  • Claude and ChatGPT wired into Microsoft 365, Google Workspace, Monday.com or your own database.
  • Voice assistants that answer spoken questions about live company data.
  • Vision pipelines that review photos or video against rules you set.
  • Custom MCP servers and API integrations, so an AI assistant can use your internal tools.

Proof

Shipped work you can check.

An AI content pipeline for EP Event Group with a default-deny privacy gate. In its live trial it sorted 113 client clips and zero private clips reached a finished reel. EP Event Group case study

A voice assistant inside P1 Finance’s executive reporting hub that answers spoken questions about live company numbers. P1 Finance case study

Around ten agents run my own company: email triage, finance tracking, content operations and client delivery. Soul47 HQ case study

How it works

→Talk

A free intro call about the job you want off your plate. You get a written recap the same day.

→Build

Fixed scope and price, agreed in writing before work starts. You see previews as it takes shape.

→Run

Handover with docs and training, plus a final check on your real files.

Questions people ask

Who does AI automation consulting in Westchester County, NY?

I do. I’m Aaryamann Kanojia, an independent AI automation consultant based in Westchester County. I audit where AI fits, then build the agents and integrations myself, so the person who advises you also writes the code.

Is it safe to put client data into ChatGPT or Claude?

It depends on the plan and the contract. Business and API plans from the major vendors come with terms that forbid training on your data, and consumer plans often do not. I set up the right plan, keep client data separated, and put a person’s review on each output.

Can an AI agent answer our customer emails?

It can draft them. I build agents that read the message, pull the facts from your systems and write a reply. A person approves each send until you decide a category is safe to automate.

What does an AI readiness audit include?

A written report: the tasks in your business where AI saves hours, what each would cost to build, what to skip, and a priority order. It starts at $1,500 and you keep the plan whether or not I build it.

We are a law or finance firm. Can we use AI and stay compliant?

Yes, with the right setup. I design for firms with confidentiality obligations: vendor contracts that forbid training on your data, segregated client data, and human review on each output. Your compliance officer or counsel signs off on the design before anything goes live.

How much does an AI agent cost to build?

Fixing an existing tool starts at $2,500. A new tool or automation runs $2,500 to $5,000, and work that spans several systems starts at $5,000. You get one fixed number in writing before I start, and you pay the second half after it works on your own files.

Tell me what’s eating your hours.

On-site in Westchester County and New York City. Remote across the United States.