Open to AI Research Collaborations & Strategic Roles

Architecting Intelligent Systems with Mathematical Precision

Developer · Graduate AI/ML Researcher · Designer · Technical Writer

Full-Stack DeveloperGraduate AI/ML ResearcherSystems & UI DesignerGraduate StudentTechnical & Research Writer

Explore Research View Projects Get in Touch

SYS.ID: ARCH-01 // RESEARCH GRADEVERIFIED CREDENTIALS

Dhaka, Bangladesh

Verified

Principal // AI & Systems

Md Tawhidul Islam

Graduate Researcher · Developer · Designer

PyTorchNext.jsTriton

Active

research-workspace

import torch
import torch.nn as nn
from triton_kernels import selective_scan_fused

class SelectiveStateSpaceBlock(nn.Module):
    """
    Hardware-fused Selective State-Space sequence model.
    Linear complexity O(N) over long-context sequence reasoning.
    """
    def __init__(self, d_model: int = 1024, d_state: int = 64):
        super().__init__()
        self.d_model = d_model
        self.d_state = d_state
        self.in_proj = nn.Linear(d_model, d_model * 2, bias=False)
        self.x_proj = nn.Linear(d_model, d_state * 2 + 1, bias=False)
        self.dt_proj = nn.Linear(1, d_model, bias=True)
        self.out_proj = nn.Linear(d_model, d_model, bias=False)

    def forward(self, u: torch.Tensor) -> torch.Tensor:
        # u: [batch, seq_len, d_model]
        batch, seq_len, _ = u.shape
        x, z = self.in_proj(u).chunk(2, dim=-1)
        
        # Continuous parameter discretization
        delta_A_B = self.x_proj(x)
        y = selective_scan_fused(x, delta_A_B, self.d_state)
        return self.out_proj(y * nn.functional.silu(z))

CUDA v12.4Memory: 4.2 GB / 24 GB

Inference Engine: Ready

Core Competencies

Technical Skills & Expertise

A snapshot of machine learning frameworks, full-stack architectures, and design systems.

DOMAIN // 01

Data Science and AI/ML

Matplotlib

1yAdvanced

Proficiency80%

Numpy

1yAdvanced

Proficiency80%

Pandas

1yAdvanced

Proficiency80%

Scikit Learn

1yIntermediate

Proficiency65%

All capabilities shownView Domain

DOMAIN // 02

Design

Adobe After Effects

3yFoundational

Proficiency45%

Adobe Illustrator

6yAdvanced

Proficiency80%

Adobe Photoshop

6yAdvanced

Proficiency80%

Adobe Premier Pro

3yIntermediate

Proficiency65%

All capabilities shownView Domain

DOMAIN // 03

Development

Django

1yFoundational

Proficiency45%

NextJs

2yFoundational

Proficiency45%

All capabilities shownView Domain

DOMAIN // 04

Others

Context Engineering

Intermediate

Proficiency65%

Git and Github

Intermediate

Proficiency65%

All capabilities shownView Domain

DOMAIN // 05

Programming

JavaScript

Foundational

Proficiency45%

Python

2yIntermediate

Proficiency65%

SQL

2yIntermediate

Proficiency65%

All capabilities shownView Domain

Active Production Toolchain & Core Stack

15 High-Impact Technologies

Matplotlib• 1yNumpy• 1yPandas• 1yScikit Learn• 1yAdobe After Effects• 3yAdobe Illustrator• 6yAdobe Photoshop• 6yAdobe Premier Pro• 3yDjango• 1yNextJs• 2yContext EngineeringGit and GithubJavaScriptPython• 2ySQL• 2y

View Full Competency Matrix

Academic Credentials

Education & Training

Formal degrees and scholarly training from premier research institutions.

EDU.01Current

MSc, Computer Science and Engineering

University of Asia Pacific

Computer Science, Data Science, Machine Learning and Deep LearningDhaka, Bangladesh2026 – Present

EDU.02

Bachelor of Computer Application

University of Mysore

Computer ScienceKarnataka, IndiaGraduated 2025

EDU.03

BA - Theology, Islamic Studies and History

Islamic Arabic University

Theology, Islamic Studies and HistoryBangladeshGraduated 2024

View Full Academic Profile

Recognition & Credentials

Awards & Certifications

Professional recognitions, academic honors, and industry-validated certifications.

CertificationCRT.01

Featured

Professional Certificate in Data Analysis and Visualization

IBM and Edx

Jan 2026

Credential VerifiedView Source

View All Honors & Certifications

Production Systems & Open Source

Featured Projects

Production applications, distributed systems, and open-source contributions.

Software Engineering

CCI Intelligent Student Recruitment & Lead Management System (CCI-iSRLMS)

The Intelligent Student Recruitment & Lead Management System is a full-stack enterprise web platform developed to digitally transform study-abroad student recruitment operations. The system combines a public student acquisition portal with an internal CRM/LMS, centralized MySQL data layer, counselor workflow management, analytics, AI-assisted consultation, secure authentication, and operational automation. Built using Next.js, React, TypeScript, Node.js and MySQL, the platform implements a custom repository-based database architecture using mysql2/promise, eliminating dependency on Prisma's production database runtime while maintaining a structured data-access abstraction. The platform provides role-based access control, JWT authentication, HTTP-only cookies, multi-tier rate limiting, automated database backup functionality, configurable recruitment pipelines, WhatsApp communication workflows, and executive analytics. The architecture is designed for deployment in constrained cPanel/CloudLinux hosting environments, making database connection reliability, resource efficiency, deployment compatibility and operational resilience important architectural considerations.

Impact:This section needs to be written carefully because the repository contains implementation and automated tests, but does not provide a formal production load-test report. Therefore, the following should be divided into verified implementation metrics and benchmarks that still require empirical measurement. Verified implementation-level metrics

Next.js 16React 19TypeScript 5.9Node.js 20++27

Read Technical Case StudySystem Architect & Developer

Explore Full Project Portfolio

Technical Dispatches & Insights