Mark Otuya

Mark Otuya

About

Detail

Senior Fullstack Ai Enginner
Africa
Contact Mark regarding: 
work
Full-time jobs
Starting at USD2.5k/month

Timeline

work
Job
school
Education
folder
Project

Résumé

Jobs verified_user 0% verified
  • Purple
    Senior Full-Stack Ai Engineer public Remote experience
    Purple
    Apr 2026 - Current (5 months)
    Own end-to-end architecture and delivery of self-hosted Node.js/TypeScript applications — managing infrastructure, networking, security, and external database configuration independently. ▸ Build scalable backend services with REST endpoints, custom authentication, input validation, and structured logging; design data models across Supabase, Postgres, and Firebase with caching strategies, implementing caching to ensure state consistency across services. ▸ Develop and maintain Wix CLI marketplace apps with dashboard extensions and admin UIs, leveraging the Wix SDK and Platform APIs to integrate deeply with Stores, Orders, and eCommerce objects. ▸ Implement complex product logic including configurable settings panels, additional fees rules,
  • J
    Frontend Engineer
    Jobin.Cloud
    Dec 2024 - May 2026 (1 year 6 months)
    As a Frontend Engineer, I utilize Claude Code and Cursor as my primary development tools, enabling me to ship production-quality modules significantly faster than traditional workflows. AI-augmented engineering is my default mode, and I have contributed to LLM-powered feature integrations, including async model response handling, streaming outputs, structured error states, and graceful degradation in production UI flows. I led the migration from Vue to React/Next.js/TypeScript, driving architecture decisions and introducing scalable patterns that were adopted team-wide, resulting in a doubling of delivery speed. Additionally, I built complex RBAC, authenticated GraphQL APIs, and multi-step user flows for a B2B SaaS product that meets Europe
  • P
    Full Stack Developer
    ProxySock
    Nov 2024 - May 2025 (7 months)
    As a Full Stack Developer, I built a complete secure multi-tenant SaaS platform independently, utilizing a robust tech stack that included Node.js/Express for the REST API, PostgreSQL for the database schema, and implemented security measures such as Role-Based Access Control (RBAC), JSON Web Token (JWT) authentication, bcrypt, and two-factor authentication (2FA). The platform achieved over 3,000 daily transactions with an impressive 99.9% uptime. During the development process, I identified a critical payment security vulnerability in LLM-generated code during the review phase, showcasing my strong judgment and attention to detail when shipping AI-assisted code in regulated environments. I employed asynchronous programming techniques throu
  • P
    Junior Developer
    Proficient Computing
    Jun 2021 - May 2022 (1 year)
    As a Junior Developer, I built both frontend and backend features for e-commerce platforms, significantly enhancing user experience and functionality. I introduced Jest and Cypress test coverage from scratch, elevating our testing coverage from 0% to 85% by identifying critical user flows and edge cases, which improved the reliability of our applications. My role involved extensive work with Application Programming Interfaces (APIs), where I utilized REST APIs hosted on cloud infrastructure, integrating them with cloud-based backend services. Although my primary focus was on frontend development, I gained valuable hands-on experience in cloud-hosted environments, API deployment, and services running on AWS and other cloud platforms. This ex
  • J
    Web developer
    Jeho films
    Aug 2019 - May 2021 (1 year 10 months)
    Built and maintained web applications for United States based media agency. Optimized server-side performance, reducing average page load times from 4s to under 2s through backend-level optimizations.
Education verified_user 0% verified
  • Edinburgh Napier University
    BSc (Hons) Computer Science
    Edinburgh Napier University
    Jan 2022 - Current (4 years 8 months)
    Microverse Full Stack Bootcamp, 1,500 hours remote pair programming. 2022-2023
Projects (professional or personal) verified_user 0% verified
  • S
    StockSense — Multi-Agent AI Stock Intelligence
    Feb 2026 - Current (7 months)
    ▸ Architecting a multi-agent system with a supervisor agent coordinating specialised sub-agents for data ingestion, signal generation, sentiment analysis, and alert delivery for US and NGX equity markets. ▸ Structured JSON communication protocol between agents with schema validation ensuring correct typed inputs regardless of upstream model behaviour — preventing silent failures in the pipeline, effectively managing independent nodes and message passing to defend against inconsistent outputs.
  • P
    ProxySock- Live B2B SaaS public Remote experience
    Feb 2026 - Current (7 months)
    Complete B2B SaaS platform built solo — subscription billing, multi-tenant PostgreSQL, RBAC, Redis caching, AWS deployment. 3,000+ daily transactions, 99.9% uptime, managing consistency, caching invalidation, and reliability trade-offs under load. Identified and mitigated a critical payment security vulnerability in LLM-generated code during the review process.
  • R
    Resume Refined, Live AI SaaS
    Jan 2024 - Dec 2024 (1 year)
    Production AI SaaS built and launched solo. Claude and OpenAI API integration with structured prompt chains, job scraper normalising across 3 HTML structures, Supabase multi-tenant backend. Real users, live in production.
  • V
    VisitAid, Real-Time Autonomous AI Agent
    Jan 2024 - Dec 2024 (1 year)
    As a VisitAid, Real-Time Autonomous AI Agent, I developed a fully autonomous agentic pipeline that streams live video via a Stream edge network. Utilizing the Vision Agents SDK, I orchestrated frames into Gemini Realtime, enabling the model to process video and audio in a single multimodal call, delivering responses without human intervention. The agent's behavior was meticulously controlled through a structured instructions file, incorporating proactive hazard detection rules, emergency mode triggers, and output format constraints, which demonstrated prompt-level governance of autonomous agent behavior. I wrote 154 lines of Python code, built and deployed the solution within a weekend, and successfully shipped it to Railway via Docker. Add
  • R
    Rora AI, Production RAG Text-to-SQL System
    Jan 2024 - Dec 2024 (1 year)
    - Developed a RAG pipeline that efficiently converts natural language to SQL, incorporating live schema injection into the system prompt, few-shot examples, and SELECT-only output guardrails to ensure structured and verifiable results. - Implemented a reasoning-before-generation technique, allowing the model to explain the existing schema it interacts with and identify edge cases prior to SQL generation, which significantly reduced hallucinations in production. - Established a structured validation pipeline featuring a benchmark regression suite of 20 question-to-SQL pairs that runs automatically after every prompt change, effectively catching output drift before it reaches users. Designed these specific test cases to probe ambiguous phrasi