Hello, I'm

Abraham Ajibade

AI Engineer ML Engineer Data Engineer

About Me

Who I Am

I design and deploy production-ready ML systems, integrate large language models (LLMs), and build reliable data pipelines for scalable machine learning workflows.

7+ Years Exp.
6 Roles Held

Core Values

Teamwork Problem Solving Communication Adaptability Curiosity Ownership Attention to Detail Continuous Learning

Beyond Work

Family Music Reading Anime Sightseeing Hiking Soccer Football Pickleball DIY

Skills

Machine Learning

PyTorch Scikit-Learn XGBoost PySpark MLlib ONNX Deep Learning NLP Computer Vision

Generative AI

Large Language Models LangChain LangGraph Prompt Engineering Retrieval Augmented Generation Semantic Kernel Vector Databases Model Context Protocol

MLOps & LLMOps

Docker Kubernetes MLflow Triton AWS SageMaker AWS Bedrock Azure Foundry Azure Machine Learning Databricks

Data Engineering

Polars Pandas Apache Spark Apache Airflow Luigi ETL/ELT SQL

CI/CD and Version Control

GitHub Actions Jenkins Drone CI ArgoCD GitHub Azure DevOps AWS CodeBuild

Infrastructure as Code and Cloud Computing

Terraform Databricks Asset Bundles Ansible Git AWS Azure GCP

Projects

FEATURED PROJECT

Retail Chat Agent Backend

Multimodal AI chat agent that understands both text and images to unlock smarter product discovery. Powered by CLIP, BLIP, and vector embeddings—served blazingly fast on AWS with FastAPI for instantly relevant results.

AWS BLIP FastAPI LangGraph LiteLLM OpenAI CLIP PyTorch Qdrant
View on GitHub
FEATURED PROJECT

Retail Chat Agent ETL

ETL powerhouse that transforms raw e-commerce catalogs into intelligent, searchable vectors—orchestrated by Apache Airflow at scale.

Airflow Python AWS S3 PostgreSQL CLIP BLIP Qdrant
View on GitHub

Human Resources Policy RAG Chatbot

Enterprise-grade RAG solution that turns static PDFs into conversational knowledge—keeping sensitive HR data behind closed doors with local LLMs and self-hosted vector search.

Databricks Docling LangChain PDFPlumber Qdrant RAG
View on GitHub

Healthcare Data Mart & ML Pipeline

HIPAA-compliant data mart and ML pipeline ingesting ~30K records per batch with a fault-tolerant architecture (under 2% failure rate).

Databricks Asset Bundles ETL/ELT PySpark Electronic Medical Records HIPAA Compliance
View on GitHub

Where I've worked

Experience

Machine Learning Engineer @ Northwestern Medicine

Chicago, IL  ·  December 2024 – Present

  • Architected and led the end-to-end deployment of a high-throughput incident classification system (6,500+ daily reports), containerizing a fine-tuned Mixtral 7x8B model via Triton Inference Server on Azure VMSS, eliminating over 90% of the manual review workload.
  • Designed and led the deployment of an event-based radiology reports classification system, enabling faster prioritization of urgent cases and mitigating potential delays and legal risks in patient care.
  • Engineered the scalable data orchestration layer on Databricks (PySpark/SQL), automating secure ingestion and HIPAA-compliant storage of patient data, enabling real-time model feedback loops and auditability.
  • Collaborated with a team of data scientists to optimize feature engineering pipelines, implementing parallel PySpark workflows that slashed feature generation latency by 92% and directly enabled faster model training and real-time inference.
DatabricksDockerLLMOpsMLflowMLOpsPySparkKubernetesTerraformAzure DevOps

Python Developer / AI Engineer @ MyCartsOnline

Lagos, Nigeria (Contract)  ·  October 2025 – February 2026

  • Designed and deployed a multimodal search platform combining BLIP-based auto-captioning with OpenCLIP (768D) embeddings to enable high-precision text-to-image and image-to-image retrieval from a product catalog.
  • Implemented a RAG layer backed by Qdrant, improving semantic matching and boosting search precision by 40% over the baseline ChromaDB implementation through optimized vector indexing and hybrid retrieval strategies.
  • Engineered a high-throughput Airflow data ingestion pipeline that automated real-time catalog synchronization for 10K+ products at 60 items/sec.
  • Built a Model Context Protocol (MCP) server exposing inventory, product, and vendor search capabilities, enabling multi-agent systems to perform real-time lookups, supplier matching, and product discovery through a unified tool layer.
LangChainOpenCLIPBLIPRAGQdrantAirflowFastAPIMLflowGithub ActionsTerraform

Data Scientist @ Blue Lambda Technologies

Atlanta, GA  ·  May 2023 – November 2024

  • Architected and scaled end-to-end ML systems across three distinct domains (real estate, retail, finance), delivering PyTorch, XGBoost, and Scikit-Learn models to production with 95%+ uptime via AWS SageMaker for scalable batch inference and automated retraining.
  • Collaborated on the development and deployment of an internal RAG platform using LangChain and Qdrant with Gemma 3.4B, resolving up to 7,000 weekly employee queries and reducing support ticket volume by 42% with 88% first-pass accuracy.
  • Owned the data engineering backbone for 15+ concurrent models; built and maintained robust ETL pipelines using PySpark on Databricks/AWS EMR, strictly enforcing data lineage and quality gates to ensure production-grade, validated datasets.
  • Instituted proactive MLOps monitoring using Arize and MLflow to detect data and concept drift; integrated auto-alerts and rollback protocols, sustaining <5% metrics degradation across models over 18-month lifecycles.
PythonMachine LearningDockerETL/ELTSQLFastAPIMLflowGithub ActionsTerraform

Graduate Research Fellow @ University of Kentucky

Lexington, Kentucky  ·  January 2021 – May 2023

  • Implemented optimization software (AIMS, R, and Python) and prescriptive analytics on resource allocation problems for bourbon manufacturing firms, culminating in published research and improving operational efficiency by 22%.
  • Conducted rigorous statistical modeling for academic research papers, utilizing R and Python to validate complex datasets and ensure 95% confidence intervals of predictions.
  • Authored research findings based on predictive modeling techniques covering market trends for industrial hemp firms.
  • Developed strong foundations in quantitative methods, econometric analysis, and data-driven decision making directly applied to real-world agricultural economics datasets.
PythonRStatistical ModelingOptimizationEconometricsPredictive Analytics

Data Scientist @ Fiverr

Lagos, Nigeria (Remote)  ·  January 2019 – November 2020

  • Delivered custom ML solutions (Scikit-learn, XGBoost) for classification and regression across diverse client portfolios, achieving 15% average R² uplift and 22% precision gain by directly aligning model metrics to business-critical KPIs.
  • Architected a real-time incremental data ingestion system using PySpark on AWS EMR, processing over 1.5M rows per run with sub-minute latency, reducing data-to-model latency by 87% and enabling live predictive scoring.
  • Refactored legacy Jupyter notebooks into modular, PEP8-compliant packages; instituted comprehensive pytest/unittest suites (unit/integration/E2E), achieving 99% test coverage and guaranteeing audit-ready client handoffs.
  • Drove model governance and transparency by integrating SHAP values and model cards directly into production APIs, empowering clients to trust and act on predictions with verifiable confidence.
PythonXGBoostSHAPPySparkScikit-LearnSQLPandas

Content & Learning Intern @ Gidi Mobile

Victoria Island, Lagos, Nigeria  ·  January 2018 – December 2018

  • Optimized mobile learning content for a WAEC prep application serving 10,000+ students, ensuring 100% alignment with national education policy and quality standards.
  • Leveraged user feedback analysis and research to drive a 15% improvement in content engagement, providing data-backed insights for iterative platform enhancements.
  • Streamlined content delivery workflows and coordinated large-scale uploads to support organizational strategies for expanding educational access.
  • Coordinated cross-functional workflows between the research and technical teams to streamline the deployment of high-priority exam prep features.
Web ContentContent ResearchEducation Policy

Academic background

Education

Master of Science @ University of Kentucky

Lexington, Kentucky  ·  2023

  • Specialized in Hemp Economics with a focus on production optimization and price analysis.
  • Applied quantitative methods, econometric modeling and statistical analysis to real-world economic datasets in international trade and hemp industries.
  • Developed strong foundations in research methodology, optimization (linear programming), and data-driven decision making.
MicroeconomicsMacroeconomicsEconometricsStatisticsQuantitative Research

Bachelor of Science @ University of Benin

Benin-City, Nigeria  ·  2017

  • Specialized in Production Economics with a focus on technical and operational efficiency.
  • Studied microeconomics, macroeconomics, and quantitative methods applied to production systems.
  • Completed coursework in macroeconomics, microeconomics, statistics, and project planning.
MacroeconomicsMicroeconomicsStatisticsPredictive AnalyticsRegression Analysis

Have a problem?
Let's solve it.

Always open to new opportunities, collaborations, and interesting challenges in artificial intelligence and machine learning.

Reach Out View Resume