Lekan SOYEWO

Software Engineer

Toronto, Canada

lekan.calekan@cognora.ca

Profile

A results-oriented Computer Science senior at York University with strong software engineering fundamentals and cutting-edge AI expertise. Proficient in Python, Java, C/C++, and JavaScript, I have developed production-ready applications that leverage machine learning and NLP to solve complex problems. My experience includes building scalable web services with serverless technologies, fine-tuning pre-trained models, and creating custom AI solutions that deliver measurable business value.

Employment History

Founder

Cognora

  • •Spearheaded strategic vision and product development for an AI-powered educational technology startup from concept to market launch.
  • •Led development and architecture decisions, building a scalable platform using React and server-less technologies that serves thousands of users.
  • •Orchestrated all aspects of technical implementation, from engineering real-time collaboration features to implementing robust subscription infrastructure and payment integrations.
  • •Drove key technical optimizations through strategic state management and caching implementations, achieving 40% improvement in application responsiveness while maintaining high user satisfaction.

Education

York University

Bachelor of Science, Computer Science

Oct 2024 – Present (Expected Graduation: 2025) | Toronto, ON

Dalhousie University

Bachelor of Science, Applied Computer Science

Aug 2021 – May 2022 | Halifax, NS

Projects

StudyBuddy

studdybuddy.ca →
  • •Full-Stack Development: Engineered a dynamic and responsive front-end using React and architected a robust back-end with Cloudflare Workers, ensuring scalability and performance under high concurrent loads.
  • •AI Integration: Designed and implemented multi-model AI integrations to enable real-time collaboration features and sophisticated document processing capabilities.
  • •Performance Optimization: Achieved a 40% improvement in system performance by implementing strategic caching mechanisms and streamlining data flow patterns.
  • •Scalability & Low Latency: Developed backend infrastructure capable of handling high-throughput concurrent AI model interactions while maintaining sub-100ms response times.
  • •Real-Time Features: Integrated collaborative tools with WebSocket-based communication for real-time interaction.
ReactCloudflare WorkersWebSocketsNode.jsPythonOpenAI/Claude APIsRedisPostgreSQLCI/CD pipelines

Multi-Modal LLM Integration Framework

github.com/Cognora/multi-llm-api-toolkit →
  • •Architected and developed a high-performance Node.js library that unifies interactions across leading AI models including Claude, GPT-4, Gemini, and Grok.
  • •Implemented stream standardization for uniform API outputs.

Sentiment Analysis of Movie Reviews

github.com/sportynest/setiment-analysis-model →
  • •Developed a machine learning pipeline for sentiment analysis on movie reviews using natural language processing techniques.
  • •Implemented tokenization, stop-word removal, stemming, and vectorization techniques (TF-IDF/Word2Vec) to preprocess raw text data.
  • •Trained models using Logistic Regression, Support Vector Machines, and ensemble methods while experimenting with LSTM and Transformer-based architectures.
  • •Achieved significant improvement in sentiment prediction accuracy.
PythonScikit-learnTensorFlowKerasPandasNLTKMatplotlibSeaborn

Skills

PythonJavaScriptTypeScriptReactJavaC/C++Data Structures & AlgorithmsMachine LearningTensorFlowScikit-learnSQLGitNode.jsPandas