Bilal Ahmad, AI agent and automation engineer

Bilal Ahmad

Cross it off.

Every ecommerce team has this list. Same jobs, every single week.

I build the AI agents and automation that cross them off, and keep crossing them off.

What’s left is the work only people can do.

Book a call See the work

Monday

  1. SupportReply to every “where’s my order?”an agent answers from your order and policy data
  2. OperationsCopy orders into the stock sheetorders sync to the sheet on their own
  3. FinanceMatch payouts to invoicespayouts matched to invoices automatically
  4. SalesChase last week’s leadsfollow-ups go out on time
  5. MarketingPull numbers for the weekly reportthe report builds itself every Monday
  6. InventoryUpdate stock on every channelone stock count across every channel
  7. YouTalk to your customers.

Less theatre. More things that work.

Four of these desks already handed a repeated job to an agent. The work still gets done every day; people just stopped doing it. Tap a tag to see which job.

  1. SalesFree hook. Yours?
  2. MarketingFree hook. Yours?

Two hooks are still free. Which job would your team hand over first?

Book a 30 minute call

Your first agent should be boring.

Not a moonshot. One job your team repeats every day, done the same way every time, so people trust it before anything grows.

Good engineering feels like one less thing to think about.

  • Learn the work before automating it.

    I started on the operations floor, so I learn the job as it is really done: real tickets, real spreadsheets, real handoffs, and the sticky note on someone’s monitor that explains everything.

  • Give the system the right context.

    The agent gets the request, the policy, the tool it needs and exactly one job. Anything that can be a plain rule stays a plain rule. Not everything needs a brain.

  • Stay for the unglamorous part.

    Retries, monitoring, the exception nobody mentioned until Thursday. I ship it, then stay and keep it running with the people who rely on it.

Systems in daily use.

Four real systems, still clocking in every day, built with the teams that use them.

I was the person doing it manually.

ZavrAIStaff

Bilal Ahmad, AI agent and automation engineer

Bilal Ahmad

Role
Automation Team Lead
Dept
Every one of them
Duties
The Monday list

On paper: Co-founder of ZavrAI. Building AI agents that run in production.

ZavrAIStaff

Bilal Ahmad

Role
The person doing it
Dept
All of them
Duties
By hand, first

Honestly: I did the Monday jobs by hand, every department’s, and learned where the work breaks.

Bilal, pronounced Bee-Laal

I started on the team doing these jobs by hand: copying, matching, chasing. So I knew which ones deserved crossing off long before I wrote any code.

Today I lead automation at Nysonian and co-founded ZavrAI. I build AI agents and automation for ecommerce teams.

These days the job is making every team’s Monday list a little shorter: something dependable and boring first, more once people trust it.

Same desk, new nameplate.

  1. Floor Manager

    ECS

    2023 to 2025

    Ran the floor. Learned where work breaks, usually at five on a Friday.

  2. Operations Executive

    Nysonian

    Early 2025

    Wrote down every manual process. The list was long.

  3. Automation Engineer

    Nysonian

    Mid 2025

    Built the automation program from scratch, starting with the most boring job.

  4. Automation Team Lead

    Nysonian

    Aug 2025 to now

    Promoted in six months. Now I own the roadmap, the team and the Monday list.

The tools follow the problem.

A customer asks

The agent reads your orders and policies, then answers.

Numbers don’t match

One place the records live, and code that checks them.

Stock is off

The store, the stock system and the sheet agree.

Everything else talks to everything

The glue between the tools your team already uses.

Questions worth answering.

Short, honest answers. Cross off the doubts before the call.

What does an AI agent and automation engineer actually do?

I build AI agents and automation that take repetitive work off ecommerce teams: customer support, operations, finance, sales, marketing and inventory. I connect your tools, give the agent the context it needs, and keep it running in production after the demo is over.

Should my first AI agent be big and ambitious?

No. Your first agent should be boring. Start with one task your team repeats every day and make it dependable. Add complexity once people trust it. An ambitious agent nobody trusts gets switched off.

Can you automate customer support without losing the human touch?

Yes. A support agent can answer routine, low-risk requests like where-is-my-thing and how-do-I-reset questions, and close them quickly. Refunds and sensitive conversations go to a person by design. The goal is a shorter queue, and customers can still reach your team.

What tools and tech stack do you build with?

Claude API and tool calling for agents, n8n and Python for workflows, PostgreSQL for data, and integrations with Shopify, WooCommerce, Slack and your helpdesk, spreadsheet, accounting and inventory tools. I also work with RAG, Ollama and TypeScript. I pick the tool that fits the problem, not the one with the best launch video.

How long does a first automation take to go live?

A focused first automation is usually a few weeks from scoping to live in production. Retries, monitoring and the edge cases nobody mentioned are part of that work from day one, not a phase two that never comes.

Are you available to hire, and do you work remotely?

I lead automation at Nysonian and co-founded ZavrAI, and I take on selected projects and roles. I work remotely with teams anywhere. The fastest way to check fit is a short discovery call. Bring your Monday list.

Got a task your team repeats?

I’m here to

Off your list

A support inbox, a stock sheet, month-end matching, a Monday report. Drop it in the tray and we’ll take it off your team’s list.

Book a call

30 minutes. You describe the job, I tell you straight whether it’s worth automating. No deck, no hard sell.

I’m in Pakistan (PKT, UTC+5) and work remotely with teams anywhere.

Hi, thanks for scrolling all the way to Friday.

To me, good automation is quiet. It just works, every day, without anyone thinking about it.

If something here stuck with you, say hello.