AI ENGINEER · AGENTIC SYSTEMS · LLM APPLICATIONS
I'm Raynauld Ronoh, an AI engineer focused on building intelligent applications with large language models, retrieval systems, and agentic workflows.
01 / ABOUT
I work at the intersection of software engineering and applied AI—translating ambiguous problems into reliable, testable, and maintainable solutions.
My interests include AI agents, retrieval-augmented generation (RAG), local and hosted language models, semantic search, model evaluation, and the systems that help AI applications behave more predictably in real-world settings.
Portfolio note: This portfolio highlights selected projects, technical implementations, engineering decisions, and practical experience across AI and software development. Each project is intended to demonstrate my problem-solving approach, technical capabilities, and understanding of the AI development lifecycle. Project outcomes, performance metrics, and responsibilities are presented according to verified work and documented results.
02 / SELECTED WORK
Four project slots designed to demonstrate architecture, implementation, evaluation, and practical impact.
An assistant concept for searching internal knowledge, synthesizing evidence, and helping users navigate complex information with a structured agent workflow.
A semantic cache that identifies similar user questions and reuses suitable responses to reduce repeated inference and improve response speed.
A retrieval-augmented generation workflow that indexes documents, retrieves relevant context, and generates answers grounded in source material.
A lightweight evaluation pipeline for running test cases, measuring keyword coverage and latency, and surfacing failures for iterative improvement.
*Keep technologies listed only where you have implemented them. Add a GitHub link, a short demo, and measured results to each project card.
DOWNLOAD / RESUME
Download my resume as a PDF for a concise overview of my experience, technical skills, and AI engineering projects.
PDF format
03 / EXPERIENCE
Designed and refined LLM-powered workflows, including prompt strategies, retrieval steps, and tool use, to support reliable AI-assisted tasks. Evaluated model outputs against quality criteria; investigated hallucinations, inconsistent responses, and failure cases, then documented improvements. Built and maintained Python components and API integrations for AI features, with attention to error handling, logging, testing, and maintainability. Collaborated with product and engineering stakeholders to translate requirements into scoped AI solutions and communicate trade-offs in quality, latency, and cost
Developed agent workflows that decomposed user requests into steps and coordinated LLM calls, tools, and retrieval components. Integrated APIs and knowledge sources into agent workflows, including input validation, timeout handling, and graceful failure paths. Implemented retrieval-augmented generation patterns such as document chunking, embedding-based search, context assembly, and source-grounded responses where appropriate. Tested agent behavior across representative tasks, refined prompts and routing logic, and documented limitations and edge cases for maintainers
Prepared and validated datasets for AI experiments, checking data quality, formatting, missing values, and consistency before use. Supported model and prompt experiments by defining test examples, comparing outputs, recording observations, and summarizing results. Built Python scripts and reusable components for data processing, inference, and evaluation, with clear input/output handling. Assisted with integrating AI capabilities into applications and troubleshooting issues involving dependencies, API responses, and model behavior.
Implemented and maintained application features using Python and related software tools, following existing code conventions and project requirements. Developed or updated REST API endpoints and service integrations, including request validation and clear error responses where required. Investigated bugs, reproduced issues, wrote or updated unit tests, and verified fixes before changes were merged or released. Used Git-based workflows to manage changes, participate in code reviews, document technical behavior, and coordinate work with other developers
04 / TOOLKIT
A focused set of technologies for building, testing, and improving AI-powered applications. Keep this list aligned with your actual proficiency.
Ask me about my approach ↗05 / LET'S CONNECT
I'm interested in conversations about AI engineering, agentic systems, LLM applications, and opportunities to build useful technology.