🌍
ClimateNetAI
Founder and lead developer of climate-aware AI research software exploring how atmospheric conditions influence 5G wireless signal prediction.
AI research · Machine learning · Systems engineering
Olohimai Juliet Michael
AI/ML Researcher, Systems Engineer & AI Software Developer building trustworthy, climate-aware and deployable intelligent systems.
Principal Communication Engineer · NASRDASystems Engineering ResearchOpen-source AI software
RESEARCH × ENGINEERING01 / About
Research-driven. Engineering-focused.
I am a Principal Communication Engineer at NASRDA and a systems engineering researcher working at the intersection of artificial intelligence, telecommunications, atmospheric variability and intelligent infrastructure. My focus is translating research into reliable software tools that can be tested, deployed and used.
My professional experience includes satellite communications, ground-station operations and mission-control systems at Nigeria's National Space Research and Development Agency (NASRDA). I develop research software that makes predictive models more accessible and examines when their results can be trusted.
Areas of interest
Research, software engineering and operational systems are complementary parts of my work.
02 / Selected work
Featured projects
Applied AI engineering, open-source research software and reliability evaluation for real-world intelligent systems.
🛡️
ClimateNetAI-RAC 5G / RBE
Reliability-aware extensions to 5G modelling: monitor prediction quality under changing conditions, detect degradation and investigate boundaries beyond which model predictions may be unreliable.
🖥️
NARL — Standalone Streamlit App
Interactive application for multilingual AI reliability analysis. Explore NARL-60 aggregate results, upload evaluation CSV files, inspect language comparisons and export summaries. Independent beta testing is in progress.
📚
N-ATLAS Reliability Lab — Research & Toolkit
Open-source developer research toolkit led by Olohimai Juliet Michael for benchmarking N-ATLaS multilingual AI reliability. Includes NARL-60 evaluation protocols and analysis of 180 recorded model responses across four Nigerian language varieties.
▶️
N-ATLaS — Reviewer Model Demo
Separate Google Colab notebook for running genuine N-ATLaS GGUF model inference. Reviewers can execute the notebook, enter prompts and inspect generated responses and finish reasons. This demo runs in Colab, not inside the Streamlit app.
🌤️
WeatherImpactML
End-to-end one-hour temperature prediction using historical Jena weather data, chronological validation and a deployed Random Forest dashboard.
03 / Capabilities
Technical expertise
Research-to-deployment capabilities across trustworthy AI, practical software engineering and mission-critical communications systems.
🧠
AI & Machine Learning
Python, NumPy, Pandas, scikit-learn, XGBoost, TensorFlow/Keras, PyTorch; regression, tree ensembles, LSTM/GRU and time-series modelling.
🛡️
Trustworthy AI & Model Reliability
Chronological validation, distribution shift, robustness testing, prediction uncertainty, reliability monitoring, degradation detection and reliability-boundary research.
🌐
LLM Evaluation & Multilingual AI
Automated inference experiments, multilingual benchmark dataset preparation, model reliability assessment and evaluation workflows developed through N-ATLAS Reliability Lab.
💻
AI Software & Automation
Streamlit applications, Git/GitHub, Google Colab, API-based workflows, Python data processing, reproducible research pipelines and practical AI automation prototypes.
📡
Telecommunications & Space Systems
5G/6G research, RF and wireless measurements, satellite communications, ground-station operations, mission-control coordination and systems integration.
📊
Research & Experimental Methods
Longitudinal sensor data, atmospheric variability analysis, statistical modelling, comparative ML experiments, seasonal evaluation, technical reporting and research software translation.
04 / Professional background
Experience & research leadership
Engineering leadership in space communications, academic teaching and research-driven development of trustworthy AI systems.
Professional experience
Principal Communication Engineer
National Space Research and Development Agency (NASRDA) · February 2025–present
Systems engineering for satellite communications and ground-station infrastructure; technical evaluation, integration, mission-control coordination and operational reliability.
Senior Communication Engineer
NASRDA · January 2022–February 2025
Supported communication systems operations, technical assessment and reliability of engineering infrastructure.
Communication Engineer I
NASRDA · August 2018–December 2021
Communication and ground-station engineering operations, systems support and technical coordination.
Systems Analyst
Vivid Global Resources · January 2013–July 2018
Systems analysis and technology-related problem solving.
Graduate Assistant / Assistant Lecturer
Covenant University · August 2009–November 2012
Supported university teaching and academic activities in information and communication technology.
Research & software leadership
Systems engineering researcher investigating atmospheric variability, 5G signal strength prediction, temporal generalization and trustworthy machine learning.
ClimateNetAI · Project Lead & Developer
Developed open-source research software translating climate-aware wireless modelling into practical analysis tools, with RAC 5G and reliability-boundary estimation (RBE) extensions.
N-ATLAS Reliability Lab (NARL) · Team Lead
Leads development of the NARL Streamlit evaluation application, including multilingual benchmark analysis, interactive reliability dashboards and reproducible N-ATLaS inference through Google Colab.
Research translation & technical communication
Connects empirical data collection, comparative machine-learning experiments, robust evaluation and accessible research software; contributes scholarly publications and technical presentations.
Research emphasis: reliable AI under changing real-world conditions, transparent evaluation and recognition of prediction limitations.
05 / Research output
Publications & research achievements
Peer-reviewed research, accepted presentations, open-source scientific software and academic service. Earlier publications may appear under Iruemi Olohimai Juliet.
Selected research publications
Machine Learning-Based Analysis of Weather Influence on 5G Network Signal Propagation in North-Central Nigeria
Published research article · Publication DOI ↗
Impact of Some Atmospheric Parameters on 5G Network Signal in Abuja During Rainy Season Using Experimental Approach
Research article · Experimental atmospheric and wireless measurements
Assessment of Seasonal Variation of 5G Signal Strength and Signal Speed over Abuja, North-Central Nigeria
Research article · Seasonal signal strength and connectivity
Future-Proofing 5G: AI-Driven Innovations in Network Management, Edge Intelligence, and Pedagogy
Research article · AI-driven communications and network intelligence
Research software & presentations
ClimateNetAI — climate-aware and reliability-aware 5G research software
Open-source research software · MIT license · Zenodo DOI ↗ · GitHub ↗
When Simpler Models Win: Evaluating Machine Learning Generalization for Climate-Aware 5G Signal Prediction
Poster accepted · Women in Machine Learning (WiML) @ NeurIPS 2026
N-ATLAS Reliability Lab (NARL)
Ongoing research and development · Multilingual LLM evaluation and reliability benchmarking; not listed as a peer-reviewed publication
Academic service & recognition
Reviewer · ML4PS 2026ICEGOV 2025 Best Paper AwardWiML @ NeurIPS 2026 · Accepted posterOpen-source research software
Explore the full publication record and citation metrics on Google Scholar ↗. Publication authorship may use the earlier name Iruemi, O. J.
06 / Connect
Let's build useful, trustworthy AI.
Open to AI/ML engineering roles, research collaborations, applied AI projects and conversations about intelligent infrastructure, model reliability and responsible deployment.
A downloadable CV will be added when the approved document is available.