I'm Mahir Tajuar Akash, a Machine Learning Engineer at Poridhi.io, specializing in building production-ready, scalable, and reliable ML systems.
My work spans the end-to-end ML lifecycle — from data engineering and model development, to deployment, monitoring, and the cloud infrastructure that holds it all together.
Beyond ML, I actively explore the infrastructure side of AI:
- ☁️ Cloud-native architectures on AWS
- 🐳 Containerization with Docker & Kubernetes
- ⚙️ CI/CD pipelines and platform automation
- 🧠 Productionizing LLMs and RAG systems
I also bring hands-on experience as an AI/ML mentor, where my focus is teaching concepts from the fundamentals up — connecting theory with real implementation.
|
|
|
|
Machine Learning · Deep Learning · NLP · Transformers · LLMs · RAG · Fine-Tuning · LoRA / PEFT · Explainable AI
Model Deployment · Experiment Tracking · Feature Stores · Model Monitoring · Data Pipelines · Model Lifecycle Management
AWS · EC2 · VPC · Transit Gateway · S3 · EKS · Kubernetes · Docker · Docker Compose · Linux · Computer Networking
Python · FastAPI · Django · React · Git · GitHub · GitLab
Jenkins · Argo CD · Helm · Prometheus · Grafana · CI/CD Pipelines · Monitoring
Data Structures · Algorithms · OOP · Operating Systems · Computer Networks · Computer Architecture
|
A production-oriented RAG system for intelligent PDF question-answering with page-level citations. Tech: FastAPI · LangChain · FAISS · Streamlit · Docker Highlights:
|
A transformer-based NLP system for detecting hate speech in Bengali and code-mixed variants. Tech: BanglaBERT · XLM-R · LoRA · SHAP · MLflow · DVC Highlights:
|
|
Exploring secure deployment of ML models inside private AWS infrastructure — no public internet exposure. Tech & Concepts: AWS VPC · Private Subnets · Transit Gateway · EC2 · Security Groups · Network Isolation · Private ML Inference |
|
Machine Learning Engineering
│
┌─────────────────┼─────────────────┐
│ │ │
MLOps DataOps Cloud Infrastructure
│ │ │
┌─────┼─────┐ ┌─────┼─────┐ ┌─────┼─────┐
│ │ │ │ │ │ │ │ │
Deploy Track Monitor Pipe Feat Data K8s Docker AWS
│ │ │ │ │ │ │ │ │
└─────┴─────┘ └─────┴─────┘ └─────┴─────┘
│
ML Platform
Engineering
│
┌─────────┴─────────┐
│ │
CI/CD Monitoring
🚀 Goal: Build ML systems that aren't just accurate, but also deployable, reproducible, observable, scalable, and maintainable in production.
I have extensive experience as an AI/ML educator and mentor, having taught and guided learners across:
- 🐍 Python Programming
- 🧠 Machine Learning & Deep Learning
- 📚 NLP, Transformers & LLMs
- 🚀 Model Deployment (FastAPI, Docker)
- ☁️ AWS & MLOps Fundamentals
My teaching philosophy: master the fundamentals first, then connect them to real-world systems.