06/16/2026 updated


100 % available
Prompt Engineer | LangChain, OpenAI API, AI Automation
Ife, Nigeria
Worldwide
BA computer scienceApplication Programming Interfaces (APIs)Messaging Application Programming InterfaceWindows APIChatbotsLarge Language ModelsPrompt EngineeringBackendArtificial Intelligence Markup Language (AIML)GPTAutomation Anywhere
I write instructions for computer programs that use large language models, and I connect these models to software systems to automate repetitive business tasks. My main focus is making AI outputs reliable, accurate, and cost-effective.
Core Technical Skills
Prompt Engineering: Writing clear system and user instructions using methods like few-shot examples and structured data outputs (JSON) to prevent the AI from generating incorrect or fabricated information.
LangChain: Building systems that link multiple AI tasks together, setting up memory for chatbots, and connecting AI models to external data files.
OpenAI API: Integrating models like GPT-4 and GPT-4o directly into web applications, software tools, or backend scripts.
AI Automation: Using Python, Zapier, or Make.com to connect AI models to regular business tools like databases, email systems, and spreadsheets.
Specific Tasks I Handle
Fixing Unreliable AI Outputs: Taking existing prompts that give inconsistent or broken answers and rewriting them so they produce the exact output required every time.
Building AI Automation Workflows: Setting up systems where an incoming text, email, or document automatically triggers an AI model to read it, extract specific data points, and save them directly to your database.
Reducing API Token Costs: Optimizing prompt length to use fewer words so you pay less for API usage while keeping or improving the quality of the results.
Setting up Data Retrieval Systems (RAG): Using LangChain to connect AI models to your internal business documents so the system answers questions based only on your company's information.
Experience
Professional experience5 - 10 years
Freelancer since2022
Languages
GermanGoodEnglishNative speakerFrenchFluent
Project history
Wrote and tested system prompts for GPT-4 to extract specific data from raw text documents and format it strictly into JSON so it could be sent directly to the company database.
Built a retrieval-augmented generation (RAG) system using LangChain. I connected the AI model to the company's internal PDF documents so it could answer employee questions based only on provided files, eliminating incorrect answers.
Automated the customer support inbox. I connected the OpenAI API to the email system to read incoming messages, categorize the issue, and draft a direct response for the human agent to review.
Reviewed older, highly complex prompts and rewrote them to be shorter and more direct. This reduced the number of words processed per request and lowered API token costs by 25%.
Set up evaluation tests to check how often the AI gave the wrong answer. I adjusted the instruction constraints until the error rate dropped below 2%.