Ozaveshe O

Claude Multi-Agent Automation Developer

The Bodyful Soul • Freelance

Mar 2026 - Aug 2026 • 5 mos

Developed a Claude-based multi-agent automation workflow for business processes that required multiple stages of reasoning, research, validation, and decision-making rather than a single AI prompt. Designed the system to divide complex tasks into specialized stages, with individual agents responsible for planning, research, analysis, tool usage, validation, and final output generation. Created structured workflows that passed relevant information from one agent to the next, reducing the need to rely on large, unstructured prompts. Designed consistent input and output formats to ensure that each stage received the information required to complete its task accurately. Implemented validation steps that allowed agents to review intermediate results, identify missing or incomplete information, and trigger revisions when necessary. Connected agent outputs with downstream automation processes so completed tasks could move through the workflow without unnecessary manual intervention. Focused on improving reliability, consistency, and control across multi-step AI processes while maintaining human oversight at critical stages. The resulting workflow provided a structured approach to using Claude for complex automation tasks, making the process easier to monitor, refine, and scale.

Systems & Automation Specialist

Self Employed • Freelance

Mar 2026 - Jul 2026 • 4 mos

I build and maintain no-code automation systems that connect the tools a business uses to manage leads, marketing, client communication, content, and internal operations. My work focused on understanding how a process should work, mapping the required triggers and conditions, connecting the right platforms, testing the workflow, and troubleshooting issues when something does not behave as expected. I have built workflows with Make.com and n8n, using APIs, webhooks, JSON mapping, and conditional logic to move information between different systems. This includes lead intake and follow-up processes, CRM automation, data processing, notifications, content workflows, and integrations between business applications. I have worked with platforms such as GoHighLevel, ActiveCampaign, HubSpot, Pipedrive, Zoho, Google Sheets, Airtable, Supabase, Slack, Notion, ClickUp, and Twilio. A major part of my approach is using AI as a technical troubleshooting assistant. I use Claude and OpenAI to investigate errors, understand documentation, refine prompts, review workflow logic, and work through possible solutions before testing changes. Rather than simply connecting two apps, I look at the full process and identify where information enters the system, how it should be processed, where it needs to go, and what should happen when conditions change or an error occurs. I have also worked on content automation, including workflows that process content from Google Sheets and connect with tools such as Claude and Blotato for multi-platform publishing. These projects have strengthened my ability to work with structured data, external services, API connections, and multi-step workflows. My goal with each system is to make the process easier to operate and maintain. I document workflow logic, organize the steps clearly, and build with future troubleshooting in mind so the automation can be understood, tested, and improved instead of becoming another source of manual work.

claude content specialist

Clean Creations • Freelance

Apr 2026 - Jun 2026 • 2 mos

Developed Claude-based content production systems for creating social media posts, marketing copy, short-form video scripts, email campaigns, and other recurring marketing content. Designed structured prompts that guided Claude to generate content according to specific brand guidelines, target audiences, campaign goals, tone of voice, and content formats. Created reusable prompt templates and content workflows that reduced repetitive manual work and made it easier to generate consistent content across multiple channels. Refined Claude outputs to improve clarity, messaging, relevance, tone, and overall content quality while ensuring that generated content remained aligned with the intended brand voice. Built a review and refinement process that allowed AI-generated content to be checked, edited, and approved before publication. This helped reduce inconsistent messaging, irrelevant outputs, and content that did not meet the required quality standards. Organized repeatable workflows covering content ideation, prompt execution, drafting, refinement, review, and final approval. Continuously improved prompts and workflows based on content performance and output quality, making the system more reliable for ongoing content production. The resulting process helped turn content creation into a structured and repeatable workflow, enabling faster content production while maintaining human oversight, brand consistency, and control over the final published content.