Samuel Jaja profile

Samuel Jaja

AI/ML Software Engineer | Multi-Cloud(AWS First)

AI/ML Systems Engineer and Full-Stack Developer with 5+ years of experience building intelligent systems that combine data engineering, machine learning, and cloud infrastructure. Skilled in LLM fine-tuning, Retrieval-Augmented Generation (RAG), and agentic AI using AWS (Lambda, SageMaker, Bedrock, Terraform). Experienced in distributed data processing with Apache Spark and PySpark on Databricks for scalable ML workflows and data lakehouse architectures, with cross-cloud experience in Azure and GCP. Strong background in Python, .NET, FastAPI, and React/Node for developing robust APIs and AI- driven applications. Passionate about creating production-ready, explainable, and compliant AI systems that improve human-computer interaction.

Featured Projects

A selection of production AI systems, machine learning projects, and data science work. View all repositories on GitHub (SWE) and GitHub (AI/ML/Cloud/Data Science/Engineering).

RevGeni CRM – AI-Powered Sales Pipeline

Built full-stack enterprise SaaS CRM with Next.js 16, TypeScript, PostgreSQL, and GraphQL API serving AI-driven lead discovery, automated email sequences, and pipeline analytics. Integrated GPT-4 and Exa AI for intelligent company search, flexible multi-LLM architecture, and sub-5s AI response times.

Next.js 16TypeScriptPrisma ORMGraphQLPostgreSQLGPT-4ExaLangfuse
Learn more

PRIME – Multi-Agent Financial Advisor

Enterprise-grade, fully-serverless multi-agent AI platform for autonomous financial investment analysis. Orchestrates 5 agents via SQS and Lambda, distributed cross-region Bedrock setup, RAG pipeline with SageMaker embeddings, and full SaaS UI. Provisioned with Terraform, Langfuse, and Cloudwatch observability.

PythonFastAPIAWS BedrockLambdaSageMakerTerraformDockerNext.js
Learn more

StructureGPT – RAG System for Building Regulations

Fine-tuned LLaMA-3.1-8b using LoRA and 8-bit quantization for UK Building Regulations compliance. Deployed to production on Hugging Face L4-GPU at $0.8/active hour with RAGAS evaluation.

PythonLLaMALoRAChromaDBLangChain
Learn more

AI Digital Twin – Production Serverless Architecture

Built full-stack production AI system using serverless AWS (Lambda, API Gateway, Bedrock, CloudFront). Achieved <200ms global response times with Terraform IaC and automated CI/CD.

FastAPIAWS BedrockTerraformNext.js
Learn more

UK Road Traffic Accident Analysis - Big Data & Mining Project

Led an in-depth analysis of UK road traffic accident data using advanced data mining techniques to identify patterns, predict future accident occurrences, and provide actionable safety recommendations. This comprehensive project combined temporal analysis, geographical clustering, association rule mining, and time series forecasting.

PythonBERTTransformersPyTorchPandasNumpy
Learn more

News Article Classification - Machine Learning & Deep Learning

I implemented and compared multiple text classification models for categorizing news articles into World, Sports, Business, and Sci/Tech categories. Developed a comprehensive solution using traditional machine learning approaches (Naive Bayes, Random Forest) and deep learning models (LSTM, Bi-LSTM).

PythonTransformersTensorFlowRandom Forestscikit-learn NLTK
Learn more

About Me

Experience

AI Associate

GW Power-Safe, UK | Current

Design and implement AI-driven solutions to enhance company workflows by streamlining procedures, integrating systems, and automating processes. Support the testing, evaluation, and refinement of AI agents to strengthen risk identification and management. Maintain and enhance reporting tools, ensuring accuracy, usability, and efficiency. Collaborate with non-technical staff to understand business requirements and translate them into AI-enabled workflows and solutions.

Full-stack | R&D Software Engineer

CypherCrescent, Nigeria | Jan 2022 - Dec 2024

Developed enterprise-grade backend services with 40% cost savings. Enabled 70% platform growth through efficient architectures. Event-Driven Design: Implemented CQRS pattern with MediatR for command/query separation and event notifications, enabling loosely coupled microservices communication via Pub/Sub messaging patterns. Azure DevOps Integration.

Software Engineering Intern (ML & Software Development)

Integral Computing & Research Centre | June 2019 - Dec 2019

Built computation tools using C# for engineering tasks and learnt the fundamentals of both Machine Learning and Software Development.

Education & Certifications

MSc. Artificial Intelligence & Data Science (Distinction)

University of Hull, UK

Bachelor of Chemical Engineering (B.Eng)

University of Port Harcourt, Nigeria

Certifications:

  • AWS Certified AI Practitioner
  • AWS Certified Generative AI Professional (Exp Dec 2025)
  • Databricks Fundamentals
  • Databricks Generative AI Fundamentals

Technical Skills

PythonFastAPIAWSAzureLangChainRAG SystemsLLM Fine-tuningTerraformNext.jsC# .NETTypeScriptDatabricksDelta LakePyTorchTensorFlowMulti-Cloud (AWS First)Docker