Muhammad Ahmad Amjad

Muhammad Ahmad Amjad

AWS Certified Solutions Architect and DevOps Engineer specializing in MLOps and Cloud Modernization.

Based in: Topi, KPMost recently: DevOps Intern (Cloud Modernization Services), Systems Limited

Hire me

A Computer Science student at GIKI with a 3.9 CGPA and professional experience in migrating enterprise CI/CD pipelines to Azure DevOps. He has a proven track record in building end-to-end MLOps pipelines and autonomous AI agents using React, FastAPI, and AWS.

Experience

DevOps Intern (Cloud Modernization Services)

Systems LimitedLahore, PunjabinternshipJul 2026 – Aug 2026

Summary

Modernized client CI/CD infrastructure by migrating 20+ legacy continuous integration pipelines from TFS and continuous deployment from Octopus Deploy to a unified Azure DevOps Services architecture. Executed the migration of version control systems from TFVC to Git across 10+ enterprise repositories, streamlining repository management. Engineered a Retrieval-Augmented Generation (RAG) agent on Azure AI Foundry indexing over 100,000 lines of internal code and documentation, cutting manual lookup time by 50% for the cloud engineering team.

Results

  • Migrated 20+ legacy continuous integration pipelines to unified Azure DevOps Services architecture.
  • Migrated 10+ enterprise repositories from TFVC to Git.
  • Cut manual lookup time by 50% for the cloud engineering team via a RAG agent.
TFSOctopus DeployAzure DevOps ServicesTFVCGitRetrieval-Augmented Generation (RAG)Azure AI Foundry

Software Engineering Intern

CarbonTeqLahore, PunjabinternshipJun 2025 – Aug 2025

Summary

Engineered the official documentation platform for CarbonTeq's Functional Programming library using React, TypeScript, and Fumadocs. Built interactive features including live code testing and "scrollycoding" by integrating Sandpack and Codehike, accelerating new developer onboarding time by over 40%. Analyzed Functional Programming concepts to translate complex library logic into accessible developer guides.

Results

  • Accelerated new developer onboarding time by over 40%.
ReactTypeScriptFumadocsSandpackCodehikeFunctional Programming

Skills

Technical

PythonPython
AWS
FastAPIFastAPI
ReactReact
TypeScriptTypeScript
GitGit
RAG
REST APIs
OpenCVOpenCV
Azure AI Foundry
AI Agents
Conversational AI
API Integration
JavaScriptJavaScript
Ultralytics YOLO
Pydantic
EC2
ECS
ECR
S3
RDS
LinuxLinux
Async Programming
IAM
GitHub ActionsGitHub Actions
DockerDocker
TerraformTerraform
Microsoft Azure
SQLSQL
PostgreSQLPostgreSQL
SSH
MLflow
DVC
DagsHub
LangChain
Node.jsNode.js
Lambda
API Gateway
BashBash
SSE
Distributed Systems
TensorFlowTensorFlow
Scikit-learn
DynamoDB
ONNX
KubernetesKubernetes
n8n
TensorRT
Minikube
LoRA

Projects

CallPilot - Autonomous Voice AI for Appointment Scheduling

Full-Stack Developerpersonal
Built a voice-first AI agent using Python, FastAPI, React, and TypeScript for autonomous appointment scheduling from natural-language commands. Integrated ElevenLabs conversational AI with asynchronous workflows to contact up to 15 providers in parallel, check calendars, calculate travel time, and automate booking. Implemented real-time SSE streaming between the Python backend and React frontend with modular tools and mock fallbacks.
PythonFastAPIReactTypeScriptElevenLabsSSE

Cloud File Storage Platform

Cloud Architect / Developerpersonal
Designed a multi-user file storage platform using React, TypeScript, FastAPI, PostgreSQL, and Amazon S3, focusing on scalable cloud architecture. Implemented JWT authentication, bcrypt password hashing, route-level authorization, and strict file ownership isolation. Containerized and deployed with Docker, ECS, ECR, and RDS; automated CI/CD with GitHub Actions, Terraform IaC, OIDC authentication, and least-privilege IAM.
ReactTypeScriptFastAPIPostgreSQLAmazon S3JWTbcryptDockerECSECRRDSGitHub ActionsTerraformOIDCIAM

Roomify

personal
A platform to match different roommates based on their traits.
Scikit-learn

Cancer Detection ResNet

personal
Trained a ResNet model on the ORC dataset and performed hyperparameter tuning to improve performance and achieve higher accuracy.
TensorFlow

UAV Object Detection Training & MLOps Pipeline

MLOps Engineeracademic
Building an end-to-end MLOps pipeline for Team Foxtrot GIKI's UAV object detection system, covering synthetic dataset generation, model training, evaluation, and edge deployment. Developing a reproducible OpenCV synthetic dataset generator parameterized by shape size, blur, backgrounds, and random seed, while using Git, DVC, and DagsHub to version datasets, configurations, and model weights. Tracking experiments with MLflow and exporting YOLO models to ONNX and TensorRT for real-time inference on the NVIDIA Jetson Orin Nano.
OpenCVGitDVCDagsHubMLflowYOLOONNXTensorRTNVIDIA Jetson Orin NanoTensorFlow

Kubernetes Practice Project

personal
Built and containerized a small FastAPI application to practice Kubernetes orchestration and CI/CD setup.
KubernetesMinikubeFastAPIDocker

Hackathon Fine-tuning Project

personal
Fine-tuned a provided model during a hackathon using the LoRA technique.
LoRA

Education

Bachelor of Science

Ghulam Ishaq Khan Institute of Engineering Sciences and TechnologyComputer ScienceGPA 3.9Jun 2027

Recognition

Certifications

AWS Certified Solutions Architect - Associate (SAA-C03)

Amazon Web Services (AWS)Credential Score: 845/10002026

Deep Learning

CampusX

Machine Learning Specialization

Stanford

Deep Learning with Computer Vision

freeCodeCamp
Hire MuhammadReach out about a role, a contract or a conversation.For recruiters

Muhammad's twin is AI, it can make mistakes.

More people on Kanvis

Browse more peopleCreate your own Kanvis