Hi there! 👋 I’m Mikias, a Machine Learning Engineer, Product Manager, and Full-Stack Developer with a passion for building AI-powered products and scalable software solutions. With a strong foundation in Electrical and Computer Engineering and hands-on experience in AI/ML, product management, and software development, I thrive at the intersection of technology and innovation.

• 🧠 Machine Learning Engineer:

  • Built and deployed sentiment analysis, machine translation, and text summarization models using transformer architectures (e.g., T5, BERT), improving accuracy by 15%.
  • Developed a chatbot using the Reformer model, reducing query resolution time by 30%.
  • Designed image classification and segmentation systems for medical imaging and autonomous driving using ResNet, YOLO, and U-Net.
  • Created fraud detection models, reducing fraudulent transactions by 18%.
  • Optimized ML pipelines using TensorFlow and TensorFlow Lite, improving edge device performance by 25%.

• 📊 Product Manager:

  • Led cross-functional teams to deliver outsourcing products in the gaming industry, improving team productivity by 20%.
  • Designed and built a payment website with integrations for local payment providers (Tele Birr, Enat Bank).
  • Pioneered the Sewasew Reward System to incentivize client participation and enhance data collection.
  • Introduced agile methodologies, fostering teamwork, adaptability, and accountability across development teams.

• 💻 Full-Stack Developer:

  • Developed RESTful APIs and optimized application performance, ensuring seamless user experiences.
  • Built a Music Contest Management Platform, supporting the music industry and strengthening market positioning.
  • Created a comprehensive admin panel for managing users, payments, and agents, simplifying platform operations.

• Machine Learning:

  • Mushroom Classification using Decision Tree.
  • Image Compression with K-Means Clustering.
  • Anomaly Detection in Server Networks.
  • Movie Recommendation Systems.
  • Lunar Lander Reinforcement Learning Agent.

• Deep Learning:

  • Face Recognition and Emotion Detection in Facial Images.
  • Art Generation with Neural Style Transfer.
  • Jazz Solo Generation with LSTM.
  • Neural Machine Translation (NMT).
  • Named-Entity Recognition (NER).
  • Extractive Question Answering.

• NLP:

  • Sentiment Analysis & Word Vector Modeling.
  • Machine Translation & Document Search.
  • Auto-Correction & Language Models.
  • Question Pairing.

• Generative AI with LLMs:

  • Fine-tuned a Generative AI Model for Dialogue Summarization.
  • Fine-tuned FLAN-T5 with Reinforcement Learning for Positive Summaries.

• Programming Languages: Python, Java, C/C++, SQL, JavaScript
• Machine Learning Frameworks: TensorFlow, PyTorch, HuggingFace, Scikit-learn
• NLP Tools: Transformers, LangChain, RAG, NLTK, Spacy
• Computer Vision: OpenCV, YOLO, U-Net
• Web Development: HTML, CSS, JavaScript, React, Node.js, Django
• Cloud Platforms: AWS, Azure, GCP
• Tools: Docker, Git, Linux, Streamlit
• ML & DS Libraries: Numpy, Pandas, Matplotlib, Tensorflow, Pytorch, HuggingFace, Transformers, LangChain, RAG, Scikit-learn, NLTK, Gensim, Spacy, OpenCV, Scipy, Statsmodel, Snowflake, Power BI, Wandb, MLFlow, Streamlit, Databricks, Spark.
• ML concepts & Techniques: Natural Language Processing, Machine Translation, Generative AI, Recommender Systems, Deep Learning, Computer vision, Reinforcement Learning, Transfer Learning, Data Cleaning and Preprocessing, Feature Engineering, Multimodal Learning, Time Series Forecasting, Predictive modeling, Model Deployment, API development, Technical Writing, ETL, A\B Testing
• Product Management: Agile, Scrum, PRDs, FSDs, Product Roadmaps

• Mathematics for Machine Learning: Linear Algebra, Multivariate Calculus (Imperial College London).
• Introduction to Statistics (Stanford University).
• Essential Linear Algebra for Data Science, Algebra and Differential Calculus for Data Science (University of Colorado Boulder).
• Machine Learning, Deep Learning, DeepLearning.AI TensorFlow Developer, TensorFlow: Data and Deployment, TensorFlow: Advanced Techniques, Generative Adversarial Networks (GANs), Natural Language Processing, Machine Learning in Production (DeepLearning.AI).
• IBM AI Product Manager, Generative AI for Product Managers (IBM & SkillUp EdTech).
• Product Ideation, Design, and Management (University of Maryland, College Park).
• Generative AI for Everyone (DeepLearning.AI).

• Delivered AI-powered solutions that contributed to a 15% revenue increase.
• Scaled AI infrastructure to support rapid innovation and deployment.
• Reduced ETL time by 40% using TensorFlow Data Services.
• Increased user engagement by 25% through recommendation engines.

• Solution Mindset: Offers unique perspectives independently.
• Service Mindset: Collaborative and approachable.
• Presentation Skills: Clearly communicates complex ideas to non-technical stakeholders.
• Comprehensive Documentation: Thoroughly documents and standardizes code and processes.
• Proactive: Anticipates needs and takes initiative.
• Flexible: Adapts quickly to new situations.
• Attention to Detail: Delivers high-quality work.
• Cultural Understanding: Works effectively in diverse environments.

• 🔍 Exploring cutting-edge AI technologies like transformers, GANs, and LLMs.
• 📈 Driving product growth with data-driven insights and AI solutions.
• 🌐 Collaborating on open-source projects and contributing to the tech community.

• 📫 Reach me at: mikias27worku@gmail.com
• 💼 LinkedIn: https://www.linkedin.com/in/mikias-worku/
• 🌟 Check out my projects below!