πŸ‘‹ Hi, I'm

MD Millat Hosen

AI Researcher Γ— Web Developer Γ— Digital Marketer

Digital Marketing & SEO Intern with the International Relations Division at Sharda University. Building efficient AI systems, production websites, and data-driven marketing campaigns. Seeking Masters opportunities in AI/ML research.

01. About

I'm a B.Tech CSE student at Sharda University (graduating June 2026), working as a Digital Marketing & SEO Intern with the International Relations Division.

My work spans AI research, web development, and digital marketing. I specialize in fine-tuning large language models (Mistral-7B) using parameter-efficient techniques like LoRA and QLoRA, with a focus on deployment in resource-constrained settings. I've authored an arXiv preprint (arXiv:2504.15610) on parameter-efficient finetuning for educational guidance in low-resource environments, with 3 citations indexed on Semantic Scholar.

Previously worked as a Web Developer at Study International Admission Care (SIAC) (Feb 2023 - Mar 2024), building production websites and managing digital marketing campaigns. My technical expertise includes Python, C++, Java, web development (React/Next.js), WordPress, MySQL, cloud platforms, and server managementβ€”alongside practical experience in SEO, analytics, and performance optimization.

Currently seeking Masters opportunities in AI/ML research, particularly in areas of LLM efficiency, multi-agent systems, and AI for education. Also open to internships and collaborations where strong fundamentals and measurable impact matter.

02. Research Interests

LLM Optimization

  • β€’ Parameter-efficient fine-tuning (LoRA, QLoRA)
  • β€’ Model compression for edge deployment
  • β€’ Resource-constrained AI systems
  • β€’ Efficient inference optimization

Autonomous Agents

  • β€’ Multi-agent collaboration architectures
  • β€’ Memory systems & context persistence
  • β€’ Agent-based reasoning & planning
  • β€’ Tool use & API integration

RAG & Retrieval

  • β€’ Retrieval-augmented generation systems
  • β€’ Vector databases & semantic search
  • β€’ Contextual chunking strategies
  • β€’ Hybrid retrieval methods

AI for Education

  • β€’ Personalized learning systems
  • β€’ Educational AI for developing regions
  • β€’ Scalable EdTech platforms
  • β€’ Adaptive assessment systems

03. Writing

04. Experience

Digital Marketing & SEO Intern

Sharda University - International Relations Division

May 2025 - Present (10 months) Β· Greater Noida, India

  • β€’ Support digital outreach initiatives to enhance university's global presence
  • β€’ Manage and grow official YouTube channel "Study at Sharda University"
  • β€’ Conduct SEO and keyword research to improve online visibility
  • β€’ Develop digital marketing campaigns for international student recruitment
  • β€’ Collaborate on social media planning and performance analysis

Web Developer

Study International Admission Care (SIAC)

Feb 2023 - Mar 2024 (1 year 2 months)

  • β€’ Built and launched production website (siacabroad.com)
  • β€’ Strengthened online presence and user experience
  • β€’ Spearheaded marketing initiatives across Facebook, Google, and YouTube
  • β€’ Maintained and optimized web infrastructure

05. Technical Skills

AI/ML

PyTorchTensorFlowHugging FaceLangChainLoRAQLoRARAGVector DBs

Development

PythonTypeScriptReactNext.jsFastAPINode.jsDockerGit

Infrastructure

LinuxVPSCloudflareVercelPostgreSQLMongoDBRedisNginx

06. Featured Projects

Kitovo.app β†’

Free Receipt Generator Tool

Professional receipt generator tool with 47+ templates (rent, gas, taxi, donation receipts). 100% private (client-side processing), works offline, exports to PDF/PNG at 300 DPI. Built for small businesses, freelancers, and landlords. Features custom branding, automatic calculations, and draft saving. Achieved 90% keyword coverage and 439ms avg load time.

πŸš€ 72 pages deployed⚑ 650KB bundle optimizedπŸ”’ 100% client-side privacy
Next.jsReactTypeScriptSEOCloudflare Pages

A LoRA-Based Approach to Fine-Tuning LLMs β†’

Published Research β€’ arXiv Preprint

arXiv:2504.15610 β€” Parameter-efficient fine-tuning of Mistral-7B-Instruct for academic advising in study-abroad contexts. Achieved 52.7% reduction in training loss,92% accuracy in domain-specific recommendations, and 95% markdown formatting support. Published April 2025, indexed on NASA ADS, Semantic Scholar (3 citations),ResearchGate, Sciety, University of Warwick, and Emergent Mind.

πŸ“„ 18 pages, 6 figuresπŸ“š 3 citationsπŸŽ“ Indexed on 7+ platforms
LoRAMistral-7B4-bit QuantizationEducational AILow-resource ML

Hugging Face Profile β†’

ML Models & Datasets

Active contributor to the Hugging Face community with machine learning models, datasets, and experimentation. Focus on LLM fine-tuning workflows, dataset preparation, and model optimization for educational applications.

πŸ€— Hugging Face ProfileπŸ“Š Models & DatasetsπŸ”¬ Active contributor
Hugging FaceTransformersDatasetsModel TrainingML Workflows

LLM Fine-tuning Research

AI Research Project

Research on efficient fine-tuning of Mistral-7B using LoRA/QLoRA for educational guidance in resource-constrained environments. Focus on deployment in developing regions with limited compute infrastructure. Achieved 4-bit quantization with minimal accuracy loss, enabling inference on consumer hardware.

🧠 Mistral-7B base modelβš™οΈ LoRA r=16, Ξ±=32πŸ’Ύ 4-bit quantization
PythonPyTorchHugging FaceLoRAQLoRATransformers

Open Source Portfolio β†’

29+ Public Repositories

Diverse collection of AI tools, web applications, Chrome extensions, and automation scripts. Active contributor to AI agent communities (Moltbook, OpenClaw). Projects span machine learning, full-stack development, browser automation, and EdTech utilities.

πŸ“¦ 29+ repositories⭐ Active contributor🀝 Community engagement
TypeScriptPythonReactAI/MLOpen Source

07. Education

Bachelor of Technology (B.Tech.) in Computer Science & Engineering

Sharda University, Greater Noida, India

June 2022 - June 2026 (Expected)

  • β€’ Digital Marketing & SEO Intern at International Relations Division
  • β€’ Focus: AI/ML, LLM optimization, web development, and digital marketing
  • β€’ Relevant coursework: Machine Learning, Deep Learning, Algorithms, Data Structures
  • β€’ arXiv preprint author: Parameter-efficient finetuning for educational guidance

Higher Secondary Certificate (HSC) - Science

Aftabgonj Govt College

2019 - 2021

08. Get in Touch

I'm actively seeking Masters opportunities in AI research, particularly in LLM optimization and autonomous agents. Also open to internships, research collaborations, and EdTech projects where strong fundamentals and measurable impact matter.