Zeeshan — AI Systems Engineer

Hi, 👋Zeeshan

AI System Engineer

Career Timeline

Sep 2024 — Present
Founder & CEO
WebWiings
Founded and lead an AI consulting and automation studio. Build custom business-process automation, LLM-driven agents, and full-stack tools for clients.
Islamabad · On-site
May 2024 — Aug 2024
Digital Marketing Manager
Edu Nexus (Contract)
Oversaw digital marketing operations for an online tech academy. Strategies hit 900% ROAS; ran competitor research to sharpen targeting.
Islamabad · Hybrid
Sep 2022 — Jun 2023
Research & Development Manager
Rzin Trade Ltd
Managed trade relations and led end-to-end R&D. Used Keepa and Helium 10 to identify high-demand products, doubling monthly sales past £10,000.
Bedford, UK · Remote
Mar 2022 — Sep 2022
Supplier & Trade Manager
Rzin Trade Ltd
Built B2B relationships with distributors and brands, negotiated supplier terms, kept orders and partnerships running smoothly.
Bedford, UK · Remote

Education

Aug 2025 — Present
B.S. in Computer Science
Virtual University of Pakistan
Four-year program covering programming, data structures, algorithms, and computer systems.

Certifications

Sep 2025
Airtable Advanced Builder
Airtable
Automation, app-building, and data modeling on the Advanced Builder track.
Jul 2024
Webflow Certification
Webflow
Responsive builds, CMS, and interactions in Webflow.

Currently learning:Fine-tuning LLMs and how to train AI models

Selected Projects

2025

Procurement Automation System

End-to-end automated PO approval, invoice parsing, and QuickBooks synchronization pipeline.

65%Faster vendor onboarding & PO approval cycle.

90%Reduction in manual QuickBooks invoice sync errors.

Skills —

n8n

OpenAIQuickBooksSlackAirtable

Event Trigger

LLM / RAG Engine

Dispatch Action

Procurement Automation System Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

2025

Customer Support Intelligence Platform

Autonomous ticket categorization, sentiment analysis, and smart draft generation system.

68%Cut in support ticket resolution time with LLM routing.

85%Automated ticket triage & draft response accuracy.

Skills —

n8n

OpenAIZendeskSlackHubSpot

Event Trigger

LLM / RAG Engine

Dispatch Action

Customer Support Intelligence Platform Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

2024

High-Volume Email Processing Engine

Intelligent inbox triage parser mapping inbound client emails directly to actionable CRM records.

80%Automated inbound email triage & draft response generation.

4xFaster response dispatch for high-priority client leads.

Skills —

n8n

OpenAIGmailSupabase

Python

Event Trigger

LLM / RAG Engine

Dispatch Action

High-Volume Email Processing Engine Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

2024

Automated CRM Lead Intelligence Pipeline

Multi-source lead scraping, enrichment, and qualification pipeline syncing directly to HubSpot.

50%Increase in qualified lead conversion speed.

95%Data enrichment accuracy across LinkedIn & HubSpot.

Skills —

n8n

OpenAIHubSpotLinkedIn

Postgres

Event Trigger

LLM / RAG Engine

Dispatch Action

Automated CRM Lead Intelligence Pipeline Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

2024

Centralized Knowledge Platform

Vector search and RAG knowledge retrieval engine over internal engineering documentation.

70%Reduction in developer search time for internal docs.

92%Semantic query accuracy across 10k+ engineering docs.

Skills —Next.js

OpenAISupabase

LangChain

TypeScript

Event Trigger

LLM / RAG Engine

Dispatch Action

Centralized Knowledge Platform Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

2024

Prompt Studio Workbench

Developer testing platform for multi-model LLM benchmarking, evaluation, and versioning.

40%Faster prompt iteration & evaluation cycles.

99.9%Consistency tracking across multi-provider LLMs.

Skills —Next.js

OpenAIAnthropicSupabase

Docker

Event Trigger

LLM / RAG Engine

Dispatch Action

Prompt Studio Workbench Architecture BlueprintProduction system workflow · Auditable agent execution pipeline

View Case Study

Working process

My Approach

Discovery

Map the real process — stakeholders, constraints, and friction — before any system is proposed.

Strategy

Decide where automation earns its place, what success looks like, and what stays human-owned.

Design

Translate the process into workflows, data flows, and decision points the team can reason about.

Development

Build in focused slices with integrations that match how the business already moves work.

Testing

Pressure-test edge cases, failure modes, and handoffs so go-live is deliberate — not hopeful.

Deployment

Ship with monitoring, clear ownership, and a rollback path if reality disagrees with the plan.

Optimization

Measure outcomes against the original friction, then tighten the loop from real usage.

Working process

My Approach

01
Discovery
Map the real process — stakeholders, constraints, and friction — before any system is proposed.
02
Strategy
Decide where automation earns its place, what success looks like, and what stays human-owned.
03
Design
Translate the process into workflows, data flows, and decision points the team can reason about.
04
Development
Build in focused slices with integrations that match how the business already moves work.
05
Testing
Pressure-test edge cases, failure modes, and handoffs so go-live is deliberate — not hopeful.
06
Deployment
Ship with monitoring, clear ownership, and a rollback path if reality disagrees with the plan.
07
Optimization
Measure outcomes against the original friction, then tighten the loop from real usage.