Opeyemi Bamigbade
Profile

Opeyemi Bamigbade

AI Systems Engineer | Multimodal AI | PhD Candidate

Open to AI Systems & Platform Roles

About Me

Designing and building scalable AI systems while advancing model capabilities through research-driven implementation. Focused on architecting and operationalising modular AI solutions across model architecture, evaluation pipelines, and high-performance inference under real-world constraints.

Applied AI Research

  • Systematically evaluates model behavior, generalization boundaries, and failure dynamics.
  • Architects model-centric AI systems to expand capability with parameter-efficient adaptation.
  • Translates research advances into modular, deployable components designed for scalable integration.
  • Focus: Model Architecture · Representation Learning · Failure Diagnostics · Capability Scaling

AI Systems Engineering

  • Engineers performance-aware workflow optimized for throughput, stability, and resource efficiency.
  • Designs modular system components that enable reliable model deployment and experimentation
  • Optimizes compute, memory, and data movement under real-world constraints.
  • Focus : Inference Engineering · Training Workflows · Systems Optimization · Performance Infrastructure

Projects

VAAS system
Inference Systems Vision Transformers Interpretability

VAAS: Inference-First Vision Anomaly Scoring System

Problem: Detect semantic and structural manipulation beyond pixel-level artefacts.
System: Dual-module (ViT attention + SegFormer consistency) producing interpretable anomaly maps and continuous scores.
Deployment: Inference-first library with modular APIs for integration into inspection and forensics pipelines, with versioned Hugging Face model releases enabling reproducible loading, evaluation, and downstream deployment. Designed for reliable integration into real-world forensic system pipelines.
View Code Model
OpsMate system
Agent Systems LLM Orchestration Stateful Runtime

OpsMate: Stateful Agent Runtime and Orchestration System

Problem: Enable reliable execution of multi-step, context-aware tasks using LLM agents beyond stateless prompt-response interactions.
System: Stateful agent runtime integrating session management, structured workflows, and tool orchestration across interaction, execution, and memory layers.
Deployment: Modular backend services (FastAPI + MCP) supporting persistent sessions, deterministic workflows, and scalable tool routing for real-world agent execution.
View Code
ML Containers
ML Systems Model Serving Containerisation

ML in Containers: Reproducible ML Deployment and CI/CD System

Problem: Ensure reproducible training, consistent model serving, and reliable deployment across heterogeneous ML environments.
System: Containerised ML workflow integrating training, versioned artefacts, and TensorFlow Serving-based inference with clear separation of system layers.
Deployment: CI/CD pipelines using GitHub Actions enabling automated build, testing, and deployment with consistent execution across environments.
View Code
TDD MLOps
MLOps Testing Frameworks ML Reliability

TDD in MLOps: Test-Driven ML System Development Framework

Problem: Lack of deterministic testing in ML systems leads to brittle pipelines, silent failures, and unreliable deployment behaviour.
System: Test-driven framework defining validation layers across data, model behaviour, and pipeline execution for controlled ML development.
Deployment: Integrated testing workflows within CI/CD pipelines ensuring reproducible validation and consistent behaviour across environments.
View Code

Technical Notes

Agent Learning
Research Note Agentic Workflow LLM Systems

Agent Learning: Execution-Aware Policy Optimization for LLM Tool Systems

Reframes tool orchestration in LLM systems as a learnable policy optimized for execution cost, latency, and constraint satisfaction rather than heuristic control.

Article PDF View Code
ML Systems Note
Technical Note GCP MLOps

Building Production Machine Learning Systems

Explores deployment of scalable, production-grade ML systems using modern MLOps workflows, focusing on system reliability, pipeline design, and cloud-native integration.

Article View Code

Research Publications

VAAS paper

VAAS: Vision-Attention Anomaly Scoring for Image Manipulation Detection in Digital Forensics

Introduces Vision-Attention Anomaly Scoring (VAAS), a dual-module framework combining global attention-based anomaly estimation in Vision Transformers with patch-level self-consistency scoring for fine-grained anomaly localization.
Formalizes attention-driven representation analysis as a structured anomaly detection mechanism, advancing modular and scalable approaches to interpretable vision-based anomaly modeling.

Paper PDF Slides
CEmbed paper

Improving Image Embeddings with Colour Features in Indoor Scene Geolocation

Introduces a colour-aware representation learning architecture integrating dominant colour descriptors and multi-space colour statistics with learned visual embeddings to enhance indoor scene understanding.
Establishes explicit colour modeling as a complementary signal to learned representations and analyzes the behavior of classification and metric learning objectives under colour-augmented settings, informing architectural and training design decisions.

Paper PDF Slides
Systematic Review

Computer Vision for Multimedia Geolocation: A Systematic Literature Review

Synthesizes 123 studies on computer vision and AI approaches to multimedia geolocation, examining architectural trends, dataset design, evaluation protocols, and application contexts across visual geolocation systems.
Identifies methodological gaps, recurring limitations, and emerging design patterns, outlining open challenges and future directions with emphasis on robustness, generalization, and real-world deployment constraints.

Paper PDF

Experience

AI Engineer

Curacel

Dec 2021–2023Hybrid

  • Led the design and delivery of the company’s first production AI system for vehicle insurance claim automation, applying computer vision techniques including object detection, damage localization, and image-based classification to automate assessment workflows and downstream decision systems.
  • Engineered and operationalized DProcessor, an OCR-driven document processing pipeline that increased claims throughput by 91% while improving reliability and operational stability.
  • Architected scalable training and inference workflows spanning model development, evaluation, optimization, and cloud deployment. Designed modular ML services on AWS, built performance-aware inference pipelines, integrated vector search into decision systems, and optimized compute, memory, and data flow across the stack.
  • Established production-grade ML practices and contributed to system-level design decisions around deployment architecture and performance trade-offs.

AI & Backend Engineer

Philanthrolab

Dec 2021–2022Remote

  • Designed and deployed ML-driven personalization systems for social and human services platforms, delivering NLP-based eligibility scoring and recommendation models integrated into live application workflows.
  • Built backend services supporting model training, evaluation, and inference, implementing data pipelines, user behavior tracking systems, and API-driven microservices connecting ML systems to production applications.
  • Optimized system performance, data flow, and operational reliability across the ML lifecycle to ensure stable, production-grade deployment.

Machine learning Engineer

Clinify

May 2020–2021Remote

  • Designed and deployed machine learning systems for healthcare applications, spanning model development, evaluation, and integration within production environments.
  • Collaborated with product and engineering teams to translate research-driven concepts into reliable, deployable ML components integrated into application workflows.
  • Contributed to system-level validation of ML-backed services and supported backend integration through API-based systems, ensuring reliability across data pipelines and production workflows

Machine learning Engineer

Data Science Nigeria

June 2019 - 2021On-site

  • Developed applied ML and NLP systems including ranking models, conversational agents, and computer vision pipelines for real-world applications.
  • Designed and deployed resource-efficient models for constrained environments, including edge deployments on hardware such as Raspberry Pi under strict compute and memory limits.
  • Built modular ML services for client integration and evaluated system trade-offs across performance, accuracy, and deployment constraints in production environments.

Skills

Programming & Frameworks

  • Python · Go
  • PyTorch · TensorFlow
  • CUDA

Modeling & Learning Paradigms

  • Multimodal & Vision–Language Modeling
  • Representation & Contrastive Learning
  • Attention Mechanisms & Transformers
  • Self-Supervised & Few-Shot Learning

Distributed Training & Performance Systems

  • DeepSpeed · Distributed Training
  • Sparse Tensor Operations
  • Mixed Precision & Quantization
  • Inference optimization (ONNX · TensorRT · TVM)

Systems & Deployment

  • MCP - FastAPI - Flask
  • AWS - GCP
  • CI/CD - Git
  • Orchestrated, Reproducible Pipelines

Education

PhD in Computer Science (in progress)

South East Technology University

2022 – Present Ireland

Advances multimodal vision systems through representation learning, robustness modeling, and scalable architecture design.
Develops architectures and evaluation frameworks for indoor scene understanding, anomaly localization, and open-set detection under real-world constraints.
Focuses on generalization, model behavior analysis, and efficient training and inference strategies.

B.Sc. Systems Engineering (First-Class Honours)

University of Lagos

2014 – 2018 Nigeria

Focused on systems design and applied machine learning for real-world decision systems.
Final-year project: Developed a machine learning–driven SaaS platform for automated Gleason score prediction, covering model development, validation, and deployment within a healthcare decision-support setting.

  • President, University of Lagos Artificial Intelligence Club
  • Ambassador, Zindi Africa (Data Science Competition Platform equivalent to Kaggle)

Peer Reviewing

Peer reviewer for AI, Machine Learning, and Computer Vision venues, assessing architectural design, methodological rigor, empirical validation, and reproducibility.
Focuses on representational innovation and system-level implications with emphasis on technical soundness and practical viability.

Learn More

Consulting

Tegus - Expert Advisor

Vector Search Embedding Systems Indexing Strategies

Provides expert advisory support to startups on vector search architectures and similarity-based retrieval systems. Advises on embedding design, indexing strategies, and trade-offs across modern vector databases, balancing performance, scalability, and deployment constraints.

Learn More

MLOps Community - Technical Author

TDD Reproducible Pipelines Model Lifecycle Testing

Authors a technical series on Test-Driven Development for ML systems, outlining structured approaches to reliability, reproducibility, and lifecycle validation. Combines conceptual guidance with implementation-level examples to formalize disciplined engineering practices in modern ML workflows.

Learn More

Datakirk - Technical Instructor

Applied Machine Learning Model Development Workflows

Delivered expert-led training on data science and machine learning for technical practitioners. Covered model development workflows, applied learning paradigms, and engineering best practices for reliable and production-aware ML systems.

Learn More

🗄️ Archive

A collection of early projects, certifications, experiments, and legacy work.

Connect

Email: bamigbadeopeyemi@gmail.com

LinkedIn: Opeyemi Bamigbade

Twitter: opeyemibami

Github: opeyemibami

Scholar: Opeyemi Bamigbade

Medium: Opeyemi Bamigbade