Hire the Best Deep Learning Experts
in Bangladesh

Hello! I'm Kazi Sohan, an analytical and skilled software engineer with over five years of professional experience. I hold a Master's degree in Information Technology from Jahangirnagar University. My career is marked by positions that have leveraged my C++, Python, and Java expertise to develop advanced AI, machine learning, and automated trading systems. My ability to innovate and deliver effective solutions has been proven through my contributions to complex projects, enhancing system performance and end-user experience. Skills: ⚫ Programming Languages: Proficient in C++, Python, Java ⚫ Expert in Algorithms and Data Structures ⚫ Specializations: Desktop Applications, Microservices, AI, Machine Learning, Computer Vision, Image Processing ⚫ Tools and Platforms: MT5, MT4, Redis, Kafka, WebSocket, REST APIs, Keras, Tensorflow, Pytorch, Numpy, Pandas ⚫ Problem Solving: Strong analytical skills with a proven track record in competitive programming ⚫ and algorithm optimization ⚫ Project Management: Experienced in leading development phases, ensuring timely delivery Professional Experience: Anyconnect: Spearheaded the development and enhancement of computer vision features, integrating AI capabilities into consumer-grade camera products, improving product performance and customer satisfaction. Next Ventures: Led the backend development for integration solutions between financial trading platforms and TradingView software, significantly enhancing data throughput and system responsiveness. Achievements: We are consistently ranked in the top tiers of national and international programming contests. Recognized for advanced problem-solving skills across various coding platforms, including Codeforces, CodeChef, and HackerRank. Publications: ⚫ Innovative methods in 3D reconstruction from 2D images using machine learning. ⚫ Improved product review classification through advanced algorithms. ⚫ Efficient real-time image classification with neural networks.

A skilled Machine Learning expert specializing in NLP and computer vision with extensive experience in end-to-end workflows comprising model training, fine-tuning, and deployment. Completed over 100 projects, including real-time person detection using YOLOv8 [1], OCR solutions with Tesseract [2], and AI chat engines leveraging OpenAI technologies [3]. Proficient in Python, TensorFlow, Keras, and FastAPI, applying these technologies to build robust ML solutions for diverse applications. Machine learning and other expertise - - Python - HuggingFace/transformers, YOLOv8 - Keras, PyTorch, TensorFlow - OpenCV - SciKit Learn - FastAPI, Docker, ReactJS Types of projects I have completed in recent years - #1. Computer Vision - - OCR with Textract, Tesseract, DocTR - Real-time person detection and recognition - Image segmentation - Image depth measurements #2. Natural Language Processing/Understanding/Generation - Chat engine with HuggingFace, GPT, LangChain, OpenAI, ChromaDB - Virtual Personal Assistant Whisper, LLaMA, Tacotron - Customer Support with Whisper, GPT4, GoogleTTS, Twilio - Tweet Generator - Recommendation Systems

I am a Machine Learning Engineer with four years of experience working across deep learning research, large scale AI systems, and production model deployment. Over the years, I have worked extensively in medical imaging, computer vision, NLP, signal processing, and large language models, building systems that range from experimental research pipelines to deployed real world AI applications. My day to day work primarily involves Python, PyTorch, TensorFlow, Keras, HuggingFace Transformers, sentence transformers, scikit learn, OpenCV, Pandas, and NumPy. I enjoy working deeply on both the research and engineering sides of machine learning, especially problems that require understanding model behavior rather than simply applying existing architectures blindly. A large part of my background is research driven. I have authored multiple peer reviewed publications in indexed journals and IEEE conferences, including publications in Neurocomputing, Healthcare Analytics, Engineering Applications of Artificial Intelligence, Telematics and Informatics Reports, and other Elsevier and IEEE venues. My research has focused heavily on explainable AI, healthcare AI, and advanced deep learning systems. Some of my published work includes CARDxnosis, an explainable knowledge driven framework for ECG diagnosis and clinical report generation, an explainable AI system for trustworthy arrhythmia detection, a CNN RNN Attention hybrid architecture for automatic modulation classification, ensemble deep learning approaches for lung cancer detection from CT scans, and SRGAN based white blood cell image generation and classification pipelines. Alongside published work, I am currently involved in research on brain tumor segmentation, ADHD and ASD classification from brain connectome graphs, epileptic seizure prediction from EEG signals, and interpretable tabular learning using graph neural networks combined with Kolmogorov Arnold Networks. Beyond research, I have substantial hands on experience building and deploying production grade AI systems. One of my major recent projects was LaborBERT v4, a domain adaptive transformer fine tuning system processing hundreds of thousands of records through a large scale training pipeline. The project involved multiple experimental setups including contrastive learning, masked language model pretraining, temporal contrastive learning, cross attention based fusion, multi task training, and Matryoshka Representation Learning. I have also built hybrid embeddings plus LLM systems for taxonomy mapping using OpenAI embeddings alongside locally hosted LLaMA and Mistral models through Ollama. In addition, I have worked on deployed clinical AI systems and a portable on device diagnostic AI solution with embedded deep learning models for point of care inference, which gave me valuable experience in optimization, deployment constraints, inference design, and production reliability. My broader project portfolio includes vehicle detection using Mask R CNN, human activity recognition on the Kinetics 700 dataset, facial keypoint detection with MultiRes UNet, semantic segmentation pipeline redesign, Stable Diffusion based image editing workflows, toxic comment classification, RASA based conversational AI systems, and large scale scraping and automation pipelines using Playwright and Selenium. I have also worked with Flask and Django based deployment pipelines and cloud hosted ML systems. From an engineering perspective, I care strongly about clean and maintainable systems. I follow disciplined workflows involving modular code design, Git based version control, reproducible experimentation, structured evaluation, bootstrap validated metrics, and detailed documentation. I am also comfortable preparing scientific reports, research papers, and journal submissions using both LaTeX and Word. What ties all of this together is that I genuinely enjoy solving difficult technical problems, especially the kind that require balancing research depth with practical engineering constraints. I am most motivated by projects where thoughtful experimentation, careful system design, and real world usability matter equally.

I am an expert AI/ML Engineer with hands-on expertise in Machine learning, Deep Learning, Computer Vision, NLP, Generative AI, and LLMs. I leverage advanced techniques to deliver innovative and impactful AI solutions for your needs. Areas of expertise: ✅ Machine Learning (Data Analysis and Visualization, Classification, Regression models, Predicting with Supervised, Unsupervised Techniques) ✅ Deep Learning (Deep Neural Networks, Transfer Learning, fine-tuning models, XAI, optimization, exploring models) ✅ Computer Vision and Image Processing (Object detection/segmentation/tracking, Image/Video classification, Image generation, OCR) ✅ NLP (Text classification, Topic modeling, NER, Semantic similarity, Text generation, Question Answering) ✅ Generative AI (Fine-tuning LLMs, RAG, Prompt Engineering, embeddings, vector database, Chatbots and AI Assistants) ✅ Audio Data Analysis (Audio classification, Sound segmentation, TTS, STT) Tech Stack: ✅ ML/DL tools: TensorFlow, Keras, PyTorch, Scikit-Learn, XGBoost, LightGBM, CatBoost ✅ NLP: LangChain, OpenAI(GPT-4, Embeddings, ChatGPT), Local LLMs, Huggingface, Gemini, NLTK, Textblob ✅ Computer Vision: Ultralytics, OpenCV, Pillow, Scikit Image, YOLO, Detectron2, OpenPose ✅ Data Analysis/Visualization Tools: Pandas, NumPy, Matplotlib, Seaborn, Plotly, Altair ✅ Cloud services: AWS, Google Cloud, Azure ✅ Project Management and Version Control: Git, GitHub ✅ MLOps: MLFlow, Docker, Github actions, Airflow, DVC ✅ Deployment: StreamLit, Flask, FastAPI ✅ Languages: Python My commitment to excellence and continuous learning ensures that I stay at the forefront of AI/ML advancements. Let’s connect to explore how I can contribute to your AI/ML projects and help you achieve your business goals with cutting-edge solutions.

Top Rated full-stack AI engineer (68 jobs, 100% JSS, 2,000+ hours on Upwork) with 6+ years building production AI systems for manufacturing, FMCG, and operations clients. I ship full SaaS products end-to-end — backend, frontend, AI features, deployment — not prototypes. I build the whole thing. I am a full-stack AI engineer with 6+ years of real business experience inside manufacturing, FMCG, and operations at Coca-Cola, Singer Bangladesh (Arçelik), Walton Hi-Tech. I have been on the client side of broken tools and bad data. Now I build the systems I wished existed. What makes me different from most full-stack freelancers: I can ship your React frontend, wire it to a FastAPI backend, connect it to PostgreSQL, add an AI agent or LLM feature, deploy it to AWS or Vercel — and I understand the business problem well enough to tell you when you are overbuilding. → SAAS MVP DEVELOPMENT This is my primary focus. If you are a founder who needs a working product in 2–4 weeks, this is what I deliver: → React / Next.js frontends — clean, responsive UI built for real users, not just demos. Tailwind CSS, component libraries, smooth UX. → FastAPI or Node.js backends — REST APIs with proper auth (JWT/OAuth), rate limiting, error handling, and documentation your team can actually use → Database design — PostgreSQL with a schema that won’t break when you hit 10,000 users. Migrations, indexing, query optimization. → Full deployment — Vercel for frontend, AWS EC2 or Railway for backend, environment configs, CI/CD with GitHub Actions. You get a live URL, not a zip file. → Stripe integration — subscription billing, usage metering, webhook handling, customer portal. Revenue from day one. → Auth systems — user registration, login, roles and permissions, password reset, email verification. Done properly, not hacked together. → AI & AGENTIC FEATURES FOR YOUR SAAS Most SaaS products in 2025 need at least one AI feature. I build them as production-ready components, not demos: → AI agents built with LangChain and LangGraph — autonomous workflows that research, extract, decide, and deliver structured outputs inside your product → RAG document systems — users upload files, ask questions, get answers with citations. Customer support, internal knowledge bases, legal document review. → LLM-powered data extraction — turn unstructured inputs (emails, PDFs, forms) into clean structured data your database can store → Automation pipelines — Airflow or Celery workers that run background jobs, send alerts, process queues, and keep your product running while nobody is watching → DATA & ANALYTICS LAYER → Power BI and Looker dashboards embedded in your product or delivered standalone — connected to live data, auto-refreshing, built for decision-making not decoration → ETL pipelines — multi-source data ingestion into PostgreSQL or data warehouses, replacing manual work permanently → ML prediction APIs — demand forecasting, churn scoring, recommendation engines — FastAPI endpoints your product can call in real time WHO I WORK WITH Founders building a SaaS MVP who need one engineer who can own the full stack — frontend, backend, AI, and deployment — without managing 4 different freelancers. Startups that have a working product but need to add AI features, improve performance, or fix the architecture before scaling. Operations and manufacturing businesses that need internal tools, dashboards, or automation systems built properly. Small teams where budget matters — you need someone who makes good decisions independently, not someone who needs constant direction. WHAT I HAVE DELIVERED ✓ Full-stack SaaS starter — Next.js frontend, FastAPI backend, PostgreSQL, JWT auth, Stripe billing hooks, rate limiting, deployed on AWS. Used as a base for client products. ✓ RAG document assistant — React UI, FastAPI backend, LangChain + ChromaDB pipeline, deployed on Vercel + EC2. Users upload PDFs, ask questions, get cited answers. ✓ LLM automation agent — searches the web, reads documents, queries a database, writes structured reports to Slack. Replaced 3 hours of manual work per week. ✓ Commercial analytics at Coca-Cola Icecek — sales data systems driving strategic decisions in a high-volume FMCG environment ✓ Quality data systems at Singer Bangladesh

I am an experienced data annotator with a strong background in various annotation tools and platforms. With 3 years of expertise. I am proficient in • Bounding box • Polygon • Polyline • Image masking • Key point annotations. I have worked with platforms like • CVAT • Super Annotate • Labelbox • Fast Label I am a quick learner when it comes to new tools and platforms. My focus is on delivering high-quality and accurate annotations within strict timelines. I am skilled at collaborating with teams, effectively communicating project requirements, and providing valuable annotated data for machine learning and computer vision projects.