\documentclass[a4paper,11pt]{article}
\usepackage{fontawesome}
\usepackage[empty]{fullpage}
\usepackage{titlesec}
\usepackage{marvosym}
\usepackage[usenames,dvipsnames]{color}
\usepackage{enumitem}
\usepackage{microtype}
\usepackage{fancyhdr}
\usepackage{ragged2e}
\usepackage{graphicx}
\usepackage[hidelinks,pdfusetitle]{hyperref}

\pagestyle{fancy}
\fancyhf{}
\fancyfoot{}
\renewcommand{\headrulewidth}{0pt}
\renewcommand{\footrulewidth}{0pt}
\setlength{\footskip}{3.60004pt}

% Margins (standard for this template)
\addtolength{\oddsidemargin}{-0.530in}
\addtolength{\evensidemargin}{-0.375in}
\addtolength{\textwidth}{1in}
\addtolength{\topmargin}{-.45in}
\addtolength{\textheight}{1in}
\urlstyle{same}
\raggedbottom
\raggedright
\setlength{\tabcolsep}{0in}

\titleformat{\section}{
  \vspace{-4pt}\raggedright\large\bfseries
}{}{0em}{}[\color{black}\titlerule\vspace{-5pt}]

\hypersetup{
  pdftitle={CV -- Md. Mutasim Billah Abu Noman Akanda},
  pdfauthor={Md. Mutasim Billah Abu Noman Akanda},
  pdfsubject={Machine learning engineer, document AI, NLP, production ML systems},
  pdfkeywords={OCR, document layout analysis, Bengali NLP, PyTorch, LLM, RAG, FastAPI, computer vision, LiveKit, WebRTC, voice AI, healthcare agent, Piper, teaching assistant},
}

\newcommand{\resumeItem}[2]{
  \item\small{
    \textbf{#1}{: #2 \vspace{-2pt}}
  }
}
\newcommand{\resumeSubheading}[4]{
  \vspace{-1pt}\item
    \begin{tabular*}{0.97\textwidth}{l@{\extracolsep{\fill}}r}
      \textbf{#1} & #2 \\
      \textit{#3} & \textit{#4} \\
    \end{tabular*}\vspace{-5pt}
}
\newcommand{\resumeSubItem}[2]{\resumeItem{#1}{#2}\vspace{-3pt}}
\renewcommand{\labelitemii}{$\circ$}
\newcommand{\resumeSubHeadingListStart}{\begin{itemize}[leftmargin=*]}
\newcommand{\resumeSubHeadingListEnd}{\end{itemize}}
\newcommand{\resumeItemListStart}{\begin{itemize}[itemsep=2pt,parsep=0pt,topsep=3pt,leftmargin=14pt]}
\newcommand{\resumeItemListEnd}{\end{itemize}\vspace{-4pt}}

% One \item[] wraps title row + nested bullets (avoids broken list layout)
\newcommand{\projectBlockStart}[2]{%
  \item[]
  \begin{tabular*}{0.97\textwidth}{l@{\extracolsep{\fill}}r}
    \textbf{#1} & #2 \\
  \end{tabular*}\vspace{-2pt}
  \begin{itemize}[itemsep=2pt,parsep=0pt,topsep=2pt,leftmargin=14pt]
}
\newcommand{\projectBlockEnd}{\end{itemize}\vspace{4pt}}

\begin{document}

\begin{center}
    {\LARGE \textsc{Md. Mutasim Billah Abu Noman Akanda}} \\ \vspace{4pt}
    {\small \textbf{Senior Machine Learning Engineer \textbullet{} Data Scientist \textbullet{} AI Researcher}} \\ \vspace{4pt}
    % {\footnotesize \textit{Open to remote roles worldwide; willing to relocate for the right opportunity.}} \\ \vspace{6pt}
    \footnotesize
    \faPhone\ +880-1773272649 \quad
    \href{https://noman024.github.io/}{\faGlobe\ Portfolio} \quad
    \href{mailto:mutasim.billah024@gmail.com}{\faEnvelope\ Email} \quad
    \href{https://www.linkedin.com/in/noman024/}{\faLinkedin\ LinkedIn} \quad
    \href{https://github.com/noman024}{\faGithub\ GitHub}
\end{center}

\section{Professional Summary}
\resumeSubHeadingListStart
\item[] \small
ML engineer and researcher with \textbf{3+ years} shipping production systems: OCR and document AI (including low-resource languages), web-scale crawling, voice interfaces, and RAG. \textbf{2+ years} as undergraduate teaching assistant in CS/AI (labs, grading, project mentoring). Full stack from literature and experimentation to deployment (Python, PyTorch, LLM/VLM optimization and quantization, FastAPI, Docker). Peer-reviewed publications in \textbf{ACM} and \textbf{Springer}.
\resumeSubHeadingListEnd

% \section{Research Interests}
% \resumeSubHeadingListStart
% \resumeSubItem{Document AI \& computer vision}
%   {Layout analysis; printed and handwritten OCR; scene text; structure from PDFs/images (including Bengali/Bangla).}
% \resumeSubItem{Machine learning, LLMs, and NLP}
%   {Deep models for vision and text; efficient inference and quantization (incl.\ VLMs); RAG, retrieval, semantic search.}
% \resumeSubItem{ML systems \& evaluation}
%   {APIs and microservices for ML; large-scale crawling and pipelines; validation and cost--latency--accuracy tradeoffs.}
% \resumeSubHeadingListEnd

\section{Experience}
\resumeSubHeadingListStart

\resumeSubheading{Apurba Technologies Ltd.}{Dhaka, Bangladesh (on-site)}
{Senior Machine Learning Engineer}{Sep 2025 -- present}
\resumeItemListStart
    \resumeItem{Borno OCR (ICT Ministry)}{Enterprise Bengali OCR, layout analysis, PDF/image to editable DOCX.}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item Integrated \textbf{dots.ocr} for Bengali; \textbf{10{,}000+} documents/day.
        \item Qwen2.5-VL 4-bit quantization: $\sim$\textbf{4$\times$} memory reduction, $\sim$\textbf{40\%} lower infrastructure cost.
        \item Bbox alignment: \textbf{+25\%} conversion accuracy, $\sim$\textbf{80\%} fewer manual fixes.
    \end{itemize}
    \resumeItem{AI Interview Bot}{Real-time voice interview platform---LiveKit WebRTC, session-isolated RAG, automated evaluation.}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item End-to-end voice pipeline (STT/LLM/TTS): $\sim$\textbf{sub-3s} average turn latency; \textbf{4-tier} TTS fallback for $\sim$\textbf{99\%} synthesis uptime.
        \item Session-isolated RAG: \textbf{10+} concurrent interviews with independent context; \textbf{zero} cross-session data leakage.
        \item Automated evaluation replacing $\sim$\textbf{20 min} manual review per candidate; job-level analytics across \textbf{multiple roles}.
    \end{itemize}
\resumeItemListEnd

\resumeSubheading{Pixalate Inc.}{Remote (California, USA)}
{Data Scientist}{Mar 2025 -- Aug 2025}
\resumeItemListStart
    \resumeItem{Focus}{Ad Intel Crawler---AI-assisted ad-network detection at web scale.}
    \item \textbf{Impact:}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item Detection pipeline: \textbf{73\%} accuracy, \textbf{2M+} sites, \textbf{50K+} pages/day.
        \item Deterministic matchers where suitable: \textbf{\$50K+/year} compute savings.
        \item Crawler orchestration: \textbf{1M+} requests/day (async).
    \end{itemize}
\resumeItemListEnd

\resumeSubheading{Green Pants Studio}{Remote (Texas, USA)}
{AI Engineer}{Jun 2024 -- May 2025}
\resumeItemListStart
    \resumeItem{Focus}{Vendidit---eCommerce scraping, valuation, and data mapping.}
    \item \textbf{Impact:}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item Scraper: \textbf{+40\%} throughput, \textbf{100K+} products/day.
        \item OLS valuation: \textbf{92\%} accuracy; \textbf{60\%} fewer pricing errors.
        \item Dockerized services: \textbf{50\%} faster deploys; \textbf{35\%} reliability improvement.
    \end{itemize}
\resumeItemListEnd

\resumeSubheading{Apurba Technologies Ltd.}{Dhaka, Bangladesh (on-site)}
{Machine Learning Engineer}{Mar 2023 -- May 2024}
\resumeItemListStart
    \resumeItem{Focus}{Bengali OCR (ICT Ministry)---layout, recognition, scene text.}
    \item \textbf{Impact:}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item YOLOv8 DLA: \textbf{85\%+} on Bengali documents.
        \item 8-bit quantized DLA: \textbf{2$\times$} faster inference, \textbf{50\%} less memory.
        \item Co-authored \textbf{ACM SE'24} and \textbf{Springer} (ICTCS) papers.
    \end{itemize}
\resumeItemListEnd

\resumeSubHeadingListEnd

\section{Teaching Experience}
\resumeSubHeadingListStart
\resumeSubheading{University of Liberal Arts Bangladesh (ULAB)}{Dhaka, Bangladesh (on-site)}
{Teaching Assistant, Department of Computer Science and Engineering}{Feb 2021 -- Jan 2023}
\resumeItemListStart
    \resumeItem{Courses}{Introduction to Programming, Object-Oriented Programming, Artificial Intelligence, Software Engineering.}
    \item \textbf{Responsibilities:}
    \begin{itemize}[itemsep=1pt,parsep=0pt,topsep=2pt]
        \item Led lab sections and tutorials (\textbf{30+} students per course); prepared lab materials and assessed coursework.
        \item Held office hours; mentored students on programming and AI/ML course projects.
    \end{itemize}
\resumeItemListEnd
\resumeSubHeadingListEnd

\section{Technical Skills}
\resumeSubHeadingListStart
\resumeSubItem{Languages \& APIs}{Python (primary), TypeScript, SQL; REST; modular services}
\resumeSubItem{ML / AI}{PyTorch, TensorFlow, scikit-learn, Hugging Face, OpenCV; EAST/CRNN OCR; QAT; vLLM, Ollama, faster-whisper, Piper (TTS)}
\resumeSubItem{LLM / NLP}{LangChain, RAG, ChromaDB, GPT/Qwen workflows, semantic search}
\resumeSubItem{Backend \& data}{FastAPI, WebSockets, LiveKit/WebRTC, SQLite, Redis, PostgreSQL, pgvector, SQLAlchemy, Pydantic}
\resumeSubItem{Automation}{Playwright, Crawl4AI, browser-use, BeautifulSoup, Scrapy, async pipelines}
\resumeSubItem{Frontend \& DevOps}{Next.js, React, Gradio, Docker, Compose, GitHub Actions, pytest, Vitest}
\resumeSubItem{Analytics}{pandas, NumPy, Matplotlib, Seaborn, MLflow, W\&B}
\resumeSubItem{Collaboration}{Git/GitHub, Jira, Notion, Slack, Agile}
\resumeSubHeadingListEnd

% \section{Languages}
% \resumeSubHeadingListStart
% \resumeSubItem{English}{Professional proficiency (reading, writing, technical communication).}
% \resumeSubItem{Bengali (Bangla)}{Native.}
% \resumeSubHeadingListEnd

\section{Education}
\resumeSubHeadingListStart
\resumeSubheading
  {University of Liberal Arts Bangladesh (ULAB)}{Dhaka, Bangladesh}
  {B.Sc.\ Computer Science and Engineering}{Jun 2019 -- Feb 2023}
\item[] \small
  \textbf{CGPA:} 3.96/4.00. \textbf{Summa Cum Laude} (top 1\%; conferred May 2024). Vice Chancellor's Honors List; Dean's List Scholarship (3$\times$). \\
  \textbf{Coursework:} ML, AI, algorithms, software engineering, computer vision, NLP.
\resumeSubHeadingListEnd

\section{Research \& Technical Projects}
\resumeSubHeadingListStart

\projectBlockStart{Voice Healthcare Agent}{\href{https://github.com/noman024/healthcare-voice-ai-agent}{\small GitHub}}
  \item \textbf{FastAPI} API and \textbf{Next.js 14} \texttt{/call} UI: REST/WebSockets for STT/TTS and agent turns; optional \textbf{LiveKit} WebRTC worker reusing same Ollama, faster-whisper, and Piper-backed TTS.
  \item \textbf{SQLite}-backed tool execution (e.g.\ appointments); session persistence and summaries aligned across REST, WebSocket, and LiveKit paths.
  \item Optional \textbf{MuseTalk} lip-sync avatar pipeline; pytest (backend), Vitest (frontend), GitHub Actions CI.
  \item \textit{Python, TypeScript, faster-whisper, Piper, Ollama, Docker Compose (LiveKit).}
\projectBlockEnd

\projectBlockStart{SenseScan (Bengali handwritten OCR)}{\href{https://github.com/noman024/SenseScan}{\small GitHub}}
  \item EAST/LANMS detection, CRNN recognition (QAT-ready), reading-order assembly.
  \item FastAPI: \texttt{/plugin} and \texttt{/v1/ocr/handwritten}; optional Gradio; DataParallel multi-GPU.
  \item \textit{PyTorch, FastAPI, OpenCV, LANMS, Pydantic.}
\projectBlockEnd

\projectBlockStart{GradConnectAI}{\href{https://github.com/noman024/gradconnectai}{\small GitHub}}
  \item CV parsing, professor/opportunity discovery, evidence-based matching, outreach drafts.
  \item \textit{FastAPI, Next.js, TypeScript, PostgreSQL/pgvector, Playwright, Crawl4AI, Docker, GitHub Actions, Ollama/vLLM.}
\projectBlockEnd


% \projectBlockStart{AI Interview Bot (real-time voice interview platform)}{\href{https://github.com/apurbatech/interview-bot}{\small GitHub}}
%   \item LiveKit WebRTC voice pipeline: faster-whisper STT, Ollama LLM, multi-backend TTS; session-isolated concurrent interviews.
%   \item Interview engine: CV/JD parsing, ChromaDB RAG, ATS scoring, evaluation worker (LLM rubrics + deterministic fallback), transcript APIs.
%   \item Rate limiting (Redis/in-memory), prompt injection defense (turn-level), Next.js candidate UI with mic/cam pre-checks.
%   \item \textit{Python, FastAPI, LiveKit, Next.js, ChromaDB, Redis, pytest, Docker.}
% \projectBlockEnd

% \projectBlockStart{RAG local document chat}{\href{https://github.com/noman024/rag-based-local-chat-box}{\small GitHub}}
%   \item Streamlit + ChromaDB, LangChain, Ollama; \textbf{90\%+} query accuracy, sub-second local responses.
%   \item \textit{Python, Streamlit.}
% \projectBlockEnd

% \projectBlockStart{Real-time parking spot tracking}{\href{https://github.com/noman024/parking-spot-tracking}{\small GitHub}}
%   \item YOLOv9 detection/tracking; $\sim$30 FPS; \textbf{92\%+} detection accuracy.
%   \item \textit{PyTorch, OpenCV, Ultralytics.}
% \projectBlockEnd

\resumeSubHeadingListEnd

\section{Publications}
\resumeSubHeadingListStart
\resumeItem
    {\textbf{Akanda, M.B.A.N.}, Ahmed, M., Rabby, A.S.A., \& Rahman, F.}
    {\textit{Optimum Deep Learning Method for Document Layout Analysis in Low Resource Languages.} ACM Southeast Conference (ACM SE'24), 199--204. \href{https://doi.org/10.1145/3603287.3651184}{doi:10.1145/3603287.3651184}}
\resumeItem
    {\textbf{Akanda, M.B.A.N.}, Prodhan, M., Sarwar, S., Raatul, A.M., Paul, B.}
    {\textit{Voice Controlled Home Automation with Cloud-Based Environment Monitoring System.} ICTCS 2022, Lecture Notes in Networks and Systems, vol.\ 623, Springer, Singapore. \href{https://doi.org/10.1007/978-981-19-9638-2_21}{doi:10.1007/978-981-19-9638-2\_21}}
\resumeSubHeadingListEnd

\section{Honors \& Awards}
\resumeSubHeadingListStart
\resumeItem{Problem setter}{ULAB Take Off Programming Contest (Summer 2022)}
\resumeItem{Vice Chancellor's Honors List}{Scholarship (Summer 2021)}
\resumeItem{2nd Runners-Up}{ULAB Take Off Programming Contest (Fall 2021)}
\resumeItem{Dean's List Scholarship}{Summer 2020, Fall 2020, Spring 2021}
\resumeItem{1st Runners-Up}{ULAB Take Off Programming Contest (Spring 2021)}
\resumeSubHeadingListEnd

\end{document}