Meharban Liaquat — Senior Full-Stack & AI Engineer

Lahore, Pakistan

About

Senior Full-Stack Engineer building production systems — and extending into AI-powered products, agentic workflows, and secure delivery.

8+ years shipping Python backends, React frontends, and cloud-native platforms for government, enterprise, and international clients — including remote engagements with US-based teams. I bridge product engineering with AI integration and practical security/GRC awareness so systems are useful, reliable, and auditable, not just demo-ready.

What I've delivered

•Cut API response times 50% with an efficient response-caching strategy
•Improved data pipeline efficiency 20% through Apache Airflow optimization
•Deployed across AWS (EC2, Amplify) and Azure App Service with Docker/K8s-ready CI/CD
•Led backend teams delivering 20+ software projects on time and within budget
•Built and hardened release pipelines (AWS Amplify, Docker, Kubernetes, Nginx)
•Implemented a complete Frappe/ERPNext HRIS and an ongoing Dynamics 365 Business Central ERP rollout

Where I'm growing / applying AI

•LLM-backed features and tooling (prompt design, API integrations, evaluation loops)
•Agentic patterns and MCP-style tool interfaces for assistants that call real systems
•Production ML serving (inference APIs, FastAPI + Docker) — not just notebooks
•Responsible use: logging, access control, and data handling for AI features

Security & GRC angle

•Hands-on familiarity with GRC tooling and compliance-minded delivery
•Secure SDLC habits: least privilege, secrets hygiene, audit-friendly cloud setups
•Comfortable partnering with security/compliance stakeholders without blocking velocity

Why this direction

I started in full-stack delivery — Django/Flask APIs, React frontends, DevOps on AWS/Azure — across government, healthcare, InsurTech, and marketplace platforms. Building CyberVectra (an AI-powered GRC platform) pulled me deeper into applying LLMs to real operational problems: risk scoring, control mapping, and compliance workflows that have to be auditable, not just clever. My personal AI/Python projects (RAG search, LangGraph agents, tool-calling assistants) are where I keep that edge sharp between client engagements.

Meharban Liaquat

System Architecture

The shape I build, repeatedly

Same request path across a GRC platform, an ERP rollout, and healthcare APIs.

request path — a shape I've shipped repeatedly

CLIENT TIER

React / Next.js

TypeScript UI, RBAC-gated views

API TIER

Django / FastAPI

REST + async endpoints, permission guards

DATA TIER

PostgreSQL + pgvector

Tenant-scoped rows, vector search

AI TIER

Agents / LLM tooling

RAG, tool-calling, evaluation loops

Skills

What I work with

Self-assessed depth, not a certification — happy to go deeper in conversation.

Full-Stack

8+ years, primary craft

95%

PythonDjangoFastAPIFlaskReactTypeScriptJavaScriptREST APIsPostgreSQLDjango REST Framework

AI / ML

Production LLM features, not just notebooks

85%

LLM AgentsRAGPrompt EngineeringLangChain / LangGraphVector Search (pgvector, Qdrant)Machine LearningComputer VisionMCP-style Tool Interfaces

DevOps / Cloud

AWS + Azure, containerized delivery

88%

AWS (EC2, Amplify)Microsoft AzureDockerKubernetesCI/CDNginxLinuxApache AirflowKafka

Security / GRC

Built an AI-powered GRC platform from scratch

80%

Application SecurityGRC (ISO 27001 / NIST / SOC 2-style)Risk ManagementSecure SDLCVulnerability Management

ERP & Business Systems

Shipped a complete HRIS, running an ongoing dual-platform rollout

75%

Frappe / ERPNextMicrosoft Dynamics 365 Business CentralHRISCustom Doctypes & WorkflowsSystem Integration

Delivery & Leadership

20+ projects led on time & budget

90%

Team LeadershipTechnical ArchitectureAgile / ScrumProject ManagementCode Review

References

People I've worked with, on request

I don't publish client quotes without explicit permission. Happy to connect you directly with people I've delivered projects for — just ask.

Ask for a Reference