Sedem Quame Amekpewu

• Fine-tuned and deployed LLMs using RLHF, SFT, and RAG for clients like Mistral and Apple.
• Designed automated data pipelines using Python and SQL, improving efficiency by 30%.
• Integrated AI-powered reasoning APIs for enterprise workflow automation.
• Cloned various API endpoints from various service providers, such as Slack, Amazon, Jira, and PayPal, to create realistic AI inference using FastAPI.
• Generated realistic data of various service providers, and imported it into Reinforcement Learning (RL Gym) to be used by various users.
• Configured MCP servers to allow AI agents to select the appropriate endpoint to fulfill users' requests.
• Lead a team of five engineers, and frequently engage in extensive code reviews and code quality checks.

• Automated SMS and USSD workflows using Python, AWS EC2, and S3, improving uptime to 99.9%.
• Implemented CRM integrations for targeted customer engagement, improving reporting accuracy by 40%.
• Developed Node.js backends for company websites ([gh.fido.money](https://gh.fido.money), [ug.fido.money](https://ug.fido.money)).
• Built various internal-facing tools to automate and assist sales, marketing, back office, and finance operations.
• Optimized existing applications by identifying bottlenecks and implementing performance improvements.
• Refactored legacy codebases for improved maintainability, paving the way for easier future updates, and conducted thorough code reviews to identify potential areas for improvement.

• Architected and built a microservices-based trading signals platform using Python, React, AWS, and Heroku; designed and managed relational and NoSQL database schemas, ensuring data integrity, scalability, and automated backup/restore procedures.

• Implemented end-to-end security, including JWT-based authentication, encryption at rest and in transit, and role-based access controls; integrated escrow management to protect user funds, and conducted thorough security audits.

• Achieved 98% uptime by designing a resilient system architecture with fault-tolerant services, auto-scaling, and robust event-driven messaging patterns.

• Led the development of responsive web and mobile interfaces, and introduced comprehensive testing practices—unit, integration, and load tests—to verify functionality, performance, and reliability across the platform.

• I created highly specialized scrapers to scrape news from various channels, built robust, fault-tolerant scrapers with IP rotation, CAPTCHA solving, authentication handling, rate limiting, DOM parsing, and data normalization to reliably ingest data from diverse sources.
• I designed and built a responsive front-end and mobile application, and handled integrations for the app.
• Developed an AI-driven news platform using microservices, AWS Lambda, and LLMs.
• Processed over 10,000 articles daily with optimized caching and event-driven architecture, reducing latency by 35%.
• Delivered a responsive web and mobile app that seamlessly integrated with the platform’s APIs, ensuring a polished user experience, and streamlined data flow.

• Engineered CI/CD pipelines and Bitbucket hygiene plugins, improving code quality by 30%.
• Built automated error classification tools reducing debugging time by 50%.
• Created observability dashboards for 200+ developers, improving response time for critical systems.
• Contract

• Developed React frontends and integrated APIs for a rent payment platform serving over 10,000 users.
• Enhanced UI accessibility and responsiveness, improving user satisfaction.

• Developed a 3D puzzle game (1K+ downloads) using Unity and C#, incorporating procedural level generation and real-time difficulty adjustments to keep gameplay fresh and challenging.
• Implemented AI-driven enemies and NPCs using pathfinding, state machines, and behavior trees, so characters react dynamically to player actions and adapt to skill level.
• Created and optimized 3D models, textures, animations, and other assets across PC and mobile platforms, refining performance with techniques like object pooling, Level of Detail (LOD) systems, and asynchronous loading.
• Enhanced AI-enabled NPCs with advanced features, such as context-sensitive interactions, quest management, and environment awareness, contribute to a smoother, more engaging player experience.

• I taught over 50 women web development and Python for data science, increasing job placement by 70%.
• Facilitated hands-on workshops using modern development tools and software frameworks.
• Implemented innovative teaching strategies to engage diverse learning styles effectively.
• Led professional development sessions for staff on emerging technologies and instructional methods.
• Contributed to a positive learning environment by maintaining an approachable demeanor and displaying enthusiasm for technology education.
• Increased student proficiency in technology by providing hands-on demonstrations and guided practice with various software applications.

• Developed an MVP for an intercity ride-sharing platform with booking and seat selection using React and REST APIs. APIs.
• Implemented RESTful APIs to integrate front-end and back-end functionalities seamlessly.
• Used NodeJS, ORM and SQL/No-SQL to develop and manage databases.
• Enhanced user experience by developing and implementing a responsive front-end interface for web applications.
• Maintained clear documentation of system architectures, workflows, and coding standards for reference by other team members.