
3
2
Nigeria
2+ years of experience
I'm Mercy Joshua, a passionate full-stack developer with a strong interest in AI, ethical tech, and collaborative innovation. My journey in tech is driven by continuous learning, purposeful collaboration, and a desire to create tools that empower others. Iβm currently exploring AI-powered tools, large language models, and how they can be used responsibly to solve real-world problems. Open to hackathons, teamwork, and any challenge that sparks growth and creativity!

Preflight is an AI Agent QA and Reliability Harness that helps teams detect and prevent AI agent failures before they reach production. AI agents are becoming increasingly capable, performing complex tasks through tool usage, decision-making, and multi-step reasoning. However, they can fail in unpredictable ways through hallucinations, unsafe actions, incorrect tool calls, prompt injection attacks, and unexpected behavior across multi-turn conversations. Developers can connect their AI agent endpoint to Preflight, where the platform automatically generates diverse testing scenarios, including normal interactions, ambiguous requests, edge cases, and adversarial attacks. These scenarios are executed through simulated multi-turn conversations to evaluate how the agent behaves under realistic conditions. Preflight uses a dual AI judge pipeline to analyze each interaction. A fine-tuned Gemma 3 safety judge, trained on AMD infrastructure, provides domain-specific safety evaluation, while DeepSeek V4 Pro serves as an independent reasoning verifier. Together, the models evaluate task success, hallucination risks, instruction following, safety issues, and prompt injection vulnerabilities. The final verdict is determined through consensus logic: if both judges classify an interaction as unsafe, it is blocked; if both classify it as safe, it is allowed; and if they disagree, the case is flagged for human review. Preflight produces a reliability report containing an overall reliability score, failure statistics, and detailed failure transcripts. By leveraging AMD Developer Cloud and ROCm-powered infrastructure, Preflight enables scalable parallel testing and supports the fine-tuning and deployment of specialized AI evaluation models. Our vision is to provide developers and organizations with a pre-deployment reliability layer for AI agents, helping teams ship safer, more trustworthy AI systems with confidence.
13 Jul 2026

ByteBound is a proactive, AI-powered travel co-pilot designed to revolutionize the entire travel experience by acting as a personal, intelligent agent. The platform is built around a fully conversational, voice-first interface, allowing users to speak their travel plans naturally and receive spoken replies, making it accessible even when hands-free. The core of the application is a sophisticated multi-agent orchestration system that seamlessly manages a user's entire journey. It begins by using the Groq API with a Llama 3 model to understand a user's natural language request, accurately extracting complex details like destinations, dates, and budget. From there, specialized agents take over, connecting to the Amadeus API to search, confirm, and book the best available flights, hotels, and local activities like tours and attractions. To enrich the user's itinerary, the system also uses the Tavily API to provide detailed suggestions for things to do at the destination. What makes ByteBound truly unique is its deep personalization and B2C model. All user preferences discovered during conversations are stored in a Neo4j Knowledge Graph, creating a reusable profile that allows the AI to make smarter recommendations over time. Furthermore, it incorporates a user's self-reported financial rhythm to make intelligent, budget-aware suggestions. To complete the B2C platform, ByteBound includes a "Partner" program. Businesses like hotels and airlines can register on the platform, and the AI will prioritize their offerings in search results when they fit the user's criteria. These partners will have access to their own analytics dashboard to monitor their performance. To ensure a professional and complete user experience, the platform also integrates the Twilio API to send instant SMS confirmations for all bookings and provides "Add to Calendar" links for all travel events.
8 Jul 2025