Oluwamayowa A.

Key Skills

Software

Other

Google Cloud Vertex AI

Internal/Proprietary Tooling

Top Subject Matter

LLM orchestration and multi-agent agentic workflows

Model training and evaluation for classification tasks

LLM-based content generation and retrieval-augmented intelligence

Top Data Types

Image

Text

Document

Top Task Types

Function Calling

Diagnosis

Text Summarization

Freelancer Overview

Nabafat.Al Technologies | Remote — AI/ML Engineer (Multi-agent reasoning workflows and LLM orchestration). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other, Weights & Biases (W&B), and Google Cloud Vertex AI. Education includes Bachelor of Science, Obafemi Awolowo University (2023). AI-training focus includes data types such as Computer Code, Programming, and Medical and labeling workflows including Function Calling, Diagnosis, and Text Summarization.

Labeling Experience

HeySynth | Remote (Austin, USA) — Fullstack AI/ML Engineer (Human-in-the-loop agentic workflows)

Google Cloud Vertex AIFunction Calling

Architected a multi-agent operating system using LangGraph to coordinate autonomous agents for operational execution. Engineered human-in-the-loop validation flows with conservative/aggressive forecast modes to maintain 99% schema compliance. Designed reasoning workflows using CoT and MCP logic to standardize data injection and reduce hallucinations in inventory workflows. • Built multiple autonomous agents with LangGraph to reduce operational overhead. • Implemented cyclic graph state-machines for human-in-the-loop validation. • Used MCP-based architectural logic for standardized data injection. • Deployed fullstack services on Google Cloud (Vertex AI) and supported real-time reasoning visualization.

2025 - Present

InspireEdge | Remote (UK) — Fullstack AI/ML Engineer

Text Summarization

Architected and deployed production LLM-powered workflows for e-commerce SMBs, generating summaries, insights, and recommendations for non-technical users. Implemented precision prompt design and RAG-based architectures to produce explainable, high-fidelity outputs. Integrated backend AI services with Shopify and WooCommerce to deliver behavioral insights to merchant dashboards. • Built LLM workflows with prompt design and model selection. • Developed RAG-based market intelligence workflows. • Integrated FastAPI AI services with Shopify/WooCommerce. • Benchmarked and monitored models using Weights & Biases in low-latency environments.

2025 - 2025

Nabafat.Al Technologies | Remote — AI/ML Engineer (Multi-agent reasoning workflows and LLM orchestration)

OtherFunction Calling

Built multi-agent reasoning workflows using LangChain for enterprise applications, focusing on structured agentic planning rather than manual annotation. Engineered orchestration across multiple LLMs (OpenAI GPT-4, Llama 3.1, and DeepSeek) with intelligent task routing for improved accuracy and token efficiency. Delivered production-ready AI modules that transform inputs into model outputs for downstream decision-making. • Developed a “Startup Adviser” and “Personalized Learning Agent” workflow logic. • Integrated multiple model providers into a single orchestration pipeline. • Implemented task routing to optimize token usage and response quality. • Containerized/scaled AI services with Docker for consistent training/inference deployment.

2024 - 2025

Quantum Leap Limited | Remote — AI/ML Engineer (Model training and MLOps)

OtherDiagnosis

Optimized model training and inference by porting TensorFlow models to JAX and NumPy to improve training efficiency. Trained classification models for sentiment analysis and customer churn prediction to support business decisioning. Streamlined MLOps with MLflow for experiment tracking and reproducible training runs. • Reduced cloud infrastructure costs by improving training/inference economics. • Developed high-accuracy predictive models for sentiment and churn. • Implemented MLflow-based experiment tracking and data versioning. • Ensured reproducibility across training cycles for evaluation and deployment.

2023 - 2023

Education

Work History