Brian Obot

• Contributed to large-scale AI model training programs by evaluating and ranking LLM outputs (RLHF) on backend code generation tasks across Python, Django, and systems-level problems.
• Built and maintained components of backend infrastructure powering AI training pipelines, including data ingestion, validation, and review-workflow services used to produce high-quality training datasets.
• Authored gold-standard Python and Django code samples used as reference solutions for model training, applying production-grade patterns in API design, testing, and error handling.
• Identified algorithmic and reasoning failure modes in LLM-generated code and provided structured feedback that directly informed model fine-tuning iterations.
• Evaluated algorithmic correctness and time/space complexity of LLM-generated solutions across data structure and systems-level coding tasks, providing structured feedback that directly informed model fine-tuning iterations.
• Assessed data structure implementation correctness and algorithm development quality across hundreds of LLM-generated solutions, applying object-oriented design principles and time/space complexity analysis to provide structured fine-tuning feedback.

• Maintain high-performance backend services using FastAPI and Rust (Axum), achieving 99.99% system uptime and enabling secure, low-latency data access across all core applications.
• Design and deploy RESTful APIs using Rust for compute-heavy routes, improving response times by 45% in real-time metrics endpoints.
• Leverage async programming models (async/await in both Rust and Python) to handle +1,000 concurrent requests, ensuring zero-lag performance during peak traffic periods.
• Integrate PostgreSQL, Redis, and containerized environments using Docker, collaborating with DevOps on CI/CD workflows to ensure seamless deployments.
• Leverage AI tooling like Claude Code to improve development efficiency and quality assurance flow.
• Implement monitoring and observability pipelines (metrics, logging, alerting) to proactively detect and resolve production incidents, maintaining 99.99% system uptime and supporting on-call operational responsibilities.
• Conduct thorough code reviews across the engineering team and author technical documentation for system architecture, API contracts, and deployment runbooks to ensure knowledge sharing and maintainability.
• Troubleshot and resolved production incidents through observability pipelines (metrics, logging, alerting), reducing mean time to recovery and optimising infrastructure costs through right-sizing containerised workloads on AWS ECS.

• Engineered scalable and high-performance backend systems using Django and Python, optimizing API performance and ensuring seamless interactions with frontend applications. Achieved a 99.9% API uptime for core banking and eVault services.
• Integrated Kafka-based event-driven architectures to enable real-time transaction processing and notifications, reducing message processing delays by 45% and ensuring high system reliability.
• Architected and accelerated Celery-powered asynchronous task queues with Redis as the broker, improving background job execution efficiency and reducing task processing times by 60%.
• Owned the full software development lifecycle — from system design and API contract definition through development, automated testing, build pipelines, and production deployment — applying established design patterns (e.g. repository, event-driven, CQRS) to ensure maintainability and scalability.

• Adopted Laravel on the job and quickly adapted to its ecosystem, leveraging my strong Python and Django background to build scalable and efficient backend solutions for financial services.
• Led the migration of a Python-based codebase to Laravel, ensuring feature parity, improved performance, and better integration with MJD Fintech's existing infrastructure. This transition resulted in a 30% improvement in API response speed and streamlined backend operations.
• Developed and optimized high-performance financial APIs using Laravel and MySQL, implementing efficient query strategies and indexing to enhance transaction processing speeds and ensure 99.9% uptime for core fintech services.

• Designed a robust data validation framework using Pydantic, leading to a 30% decrease in data corruption incidents; streamlined the development process, allowing developers to focus on feature enhancements rather than bug fixes.
• Implemented and fine-tuned Celery asynchronous processing using Redis and RabbitMQ, ensuring efficient background job processing. This approach reduced task execution latency by 50%, enabling real-time processing for critical operations.
• Built a comprehensive testing suite with Pytest, including unit, integration, and stress tests, ensuring early bug detection and improving overall test coverage to 95%. This significantly enhanced application stability and reduced regression issues.

Related Coursework: Circuit Theory, Analog and Digital Electronics, Signals and Systems, Electromagnetic Fields, Power and Control Systems, Communication Systems, Embedded Systems, Renewable Energy, Engineering Optimization, Data Analytics.