Michael Edekin

EDEKIN MICHAEL IMOUKWEDE Lagos, NG – 200113 | - |-| Github | LinkedIN Result-oriented Data scientist specialized with 4 years of full-time working experience in AI/ML, NLP, NLU, and recommendation systems. I have expertise in engineering production-level ML solutions. SKILLS     Python, SQL, Bash Machine learning, Deep learning, Statistics, Data Analysis, Data Visualization, Algorithms, and Data Structure, Computer Vision, NLP, NLU, NLG, LLM, Big Data, Data Engineering, MLops and ML system design Spacy, NLTK, Flair, HuggingFace, Spark, HDFS, Hive, TensorFlow, Pytorch, SQL, NoSQL, Flask, Django, FastAPI, ReactJs, AWS, GCP, Git, Docker, MLflow, Kubernetes, Dialogflow, Kafka, CI/CD, Azure, Rasa Leadership, Ownership, Innovation, Public Speaking, Mentoring EXPERIENCE GritinAI Head of Operations & Technical Lead (Generative AI & Cloud)     Building enterprise agents - function calling, RAG, memory modeling, code interpreter, reflection, planning, reasoning and multi-agent. Consultancy: Worked with Horizontal IOU to build AI capabilities for multiple domains (Government, healthcare, Automobile, Insurance.) Big Win: Multiple POC’s for clients ended up getting converted to a Full-fledged Data Science Project. Currently working on different AI/ML and Gen AI Projects using Azure Cloud. Assistant Technical Lead (Generative AI)        California, US Aug 2023 – Apr 2024 Developed a recommendation system using sentence embeddings and Retrieval-Augmented Generation (RAG) techniques. Utilized Azure Cloud for scalable data storage and deployment, and leveraged Large Language Models (LLMs) such as GPT-3 or similar for generating contextually relevant recommendations. Built an analytics platform utilizing Python and SQL to enable real-time tracking and analysis of diversity procurement efforts. Utilized advanced visualization libraries (D3.js, Plotly) to build dynamic dashboards and reporting tools. Built an AI driven chatbot, utilizing TensorFlow and Spacy libraries for NLP. Integrated Transformer-based models such as BERT and GPT-3 for contextual understanding and dialogue generation. Deployed Rasa for conversational flow management and dialogue state tracking. Used Apache Kafka for real-time data streaming and communication. Implemented GitLab CI/CD pipelines and workflows for Docker builds, unit testing, metrics reporting, and merge conflict checks to support MLOps and ensure codebase stability for the team. Artificial Neurons AI Data Scientist (Recommendation System & CV)  Nov 2023 – Apr 2024 Built spark data pipeline on GCP Dataproc clusters to process ~ 2 billion JSON records to train a sentence embedding using transformers model, Pytorch and NLTK on 4 million query-product pairs. Created a pipeline to retrieve ~ 2 million product features from multiple big query tables. Utilized Hadoop, Yarn, and HDFS to construct effective ETL pipelines for data processing and analysis. Engineered a recommendation system using sentence embedding and the BM25 method, which increased clicks and purchases by 10% on offline evaluation. TalkDEI Lead AI/ML Engineer (Recommendation System & Generative AI)  Edo State, NG Apr 2024 - Present Mumbai, IN Mar 2022 – Dec 2022 Utilize transformer model to design and build a healthcare recommendation algorithm using sentence embeddings and RAG, normalizing complex services for microservices model serving and KNN query searching. Analyze patient data and discern clusters of patients with analogous health profiles.   Assisting primary & secondary tracing of medical images with Computer Vision. Using 2M+ labeled datasets to train the models, enabling the identification of specific features or patterns in images for highly accurate and detailed insights. Developed comprehensive technical documentation for all research experiments, including detailed results, analysis of current shortcomings, and system architecture diagrams for deployment services and APIs. Effectively communicated technical concepts to upper management and VPs to provide clear and concise insight into the performance and limitations of the models. ApptJobs Senior Data Scientist (Recommendation System & CV)     Built a scalable recommendation system to recommend jobs to professionals using sentence embeddings and RAG techniques, resulting in 10X time reduction in HR manual sorting process, more effective and scalable than most ATS. Performing custom OCR tasks to leverage images to text of company’s relevant documents. Built a web crawler along with the backend team Implemented GitLab CI/CD pipelines and workflows for Docker builds, unit testing, metrics reporting, and merge conflict checks to support MLOps and ensure codebase stability for the team. Freelancer (Upwork, WhatsApp) Data Science & Machine Learning Engineer   Lagos State, NG Jun 2021 – Aug 2023 San Francisco, US Mar 2019 – Jun 2021 Top Rated Plus: Highly rated freelancer status (represent top 3% globally). 750+ hours of logged in effort of working on complex Machine & Deep Learning Projects with 100% Job Success Rate. EDUCATION University of Benin Bachelor of Engineering – Civil Engineering Hamoye School of Data Science Data Science ACHIEVEMENTS  TOP 6 in Hamoye School of Data Science (Worldwide)  GCP Advanced Generative AI for Developers - Google Cloud Certification Dec 2022 GPA: 3.9/5.0 Dec 2020 Top 6 Globally