Machine Learning Engineer · Technical Writer · Speaker · Part-time Chef
I build machine learning systems and write about them. Most of my work sits at the intersection of LLM infrastructure, interpretability, and developer education — turning research and tooling into things engineers can actually use.
- Writing about LLM interpretability, agent observability, and RAG architecture
- Publishing in Towards AI, Data Science Collective, AI in Plain English, and Techloy
- Covering fintech, payments, and cross-border infrastructure on the meCash blog
- Open to collaborating on interesting projects
- Reach me at onose75@gmail.com
Built and maintain the developer documentation for meCash's cross-border payments platform: API reference, authentication, sandbox guides, and integration walkthroughs for the Ramp and Payment Links APIs.
Mintlify OpenAPI MDX Docs-as-Code
| Project | Description | Tech Stack |
|---|---|---|
| Legal AI Assistant | AI assistant for legal document analysis, summarization, and querying using RAG. | LangChain Streamlit VectorDBs LLMs |
| Gemini AI Researcher | Research agent that uses Gemini to find information and return accurate citations. | Gemini API Python LangChain Streamlit |
| Local LLM Agent w/ Monitoring | Autonomous agent on local LLMs with real-time performance and drift monitoring. | Python LangChain Ollama Prometheus |
| Company Researcher Agent | Agent that researches companies, gathers public data, and generates summary reports. | LangChain Agents SerpAPI LLMs Python |
| Advanced RAG Techniques | Implementations of advanced RAG strategies sourced from recent academic papers. | Python LangChain PyTorch VectorDBs |
| Local Text-to-Image | Locally-run text-to-image generator powered by DeepSeek-Janus. | DeepSeek Python Gradio Transformers |
| RAG Project Testbed | Experimental harness for testing and evaluating RAG techniques. | Python LangChain VectorDBs Hugging Face |
ML / AI — LLMs, RAG, agents, interpretability, computer vision, PyTorch, Hugging Face, LangChain, Ollama
MLOps — feature stores, experiment tracking, monitoring, drift detection, model observability
Infrastructure — AWS, GCP, Kubernetes, Ambassador Edge Stack, Jenkins, Pulumi, GitOps
Engineering — Python, JavaScript, Django, Flutter, HTML/CSS, Web3
Writing — technical documentation, docs-as-code, developer education, API reference
- The Complete Guide to Reading a Model's Hidden Layers with Anthropic's Jacobian Lens
- Inside the Black Box: The State of LLM Interpretability in 2026
- Anthropic Found a Global Workspace Inside Language Models — Here's What It Means
- Hallucination in Large Language Models
Writing on cross-border payments, stablecoin rails, and African fintech infrastructure for meCash — full archive at blog.me-cash.com/author/ejiro.
- Open Knowledge Format (OKF) vs Vector Databases: Why Your RAG Needs a New Knowledge Format
- DeepSeek-OCR: Solving LLM Context Limits with Optical Compression
- Vector Database: What Is It and Why You Should Know It?
- From LLM Logging to Agent Observability: How We Monitor and Evaluate AI Systems in 2026
- Why Observability Is Important for Large Language Models
- LLMs on Trial: Understanding LLM Evaluation Benchmarks
- LLMs on Trial: The African LLM Evaluation Benchmarks Edition
- ML Monitoring vs ML Observability
- How to Improve Model Performance with ML Monitoring
- How to Choose the Right Solution for Machine Learning Monitoring
- How to Automate Data Drift Thresholding in Machine Learning
- Explainability and Auditability in Machine Learning
- Build a Local Code Review & Refactoring Assistant with OpenClaw + Ollama
- The Complete Guide to Running Large Language Models Locally with Phoenix and Ollama
- Why Your 2026 AI Stack Still Needs a Feature Store
- Machine Learning as a Service
- Azure ML Alternatives for MLOps
- How to Solve Reproducibility in Machine Learning
- Best Data Lineage Tools
- Building Machine Learning Chatbots: Choose the Right Platform and Applications
- Kubernetes Development Environments: From Local to Remote
- Ambassador Edge Stack vs Traefik vs NGINX
- Using Knative and Ambassador Edge Stack to Handle Traffic
- Secrets Management with GitOps and Kubernetes
- Opta vs Pulumi
- What Is Kubernetes Lens
- The AI Product Problem
- Understanding Docs-as-Code
- Building Topical Relevance for Your SaaS Product: A Guide for Tech Teams
More on Medium, Techloy, and the meCash blog.