Skip to content
View ajegetina's full-sized avatar

Highlights

  • Pro

Block or report ajegetina

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ajegetina/README.md

Software Engineer · AI Systems

I'm a software engineer, and I work on AI systems.

I work full-stack — the interface people use, the services behind it, and the reliability that keeps them up. On top of that I build LLM integrations: orchestration layers, RAG pipelines, and application-layer AI security. I hold an MSc in Software Engineering and my graduate research at Ashesi University is on efficient AI for low-resource environments.


🟢 Current Focus

  • Day-to-day: Production software engineering — backend, multi-tenant architecture, reliability, performance.
  • Differentiator: Shipping AI into products — LLM orchestration (AWS Bedrock), RAG, prompt-injection hardening.
  • Direction (MPhil, Year 2 research): Low-resource AI — lightweight model adaptation, semi-supervised ASR for underrepresented languages, retrieval for constrained settings.

🛠️ Core Engineering

Full-Stack & Systems AI Integration Security & Robustness
React / Next.js Frontends LLM Orchestration AI Security (prompt injection)
High-Performance FastAPI APIs RAG Pipeline Design Multi-tenant Isolation
Distributed Systems & Messaging Fine-tuning & Embeddings Ethical Hacking · OWASP LLM Top 10

🏗️ Featured Project: Meridian Policy Intelligence

A full-stack RAG application for high-precision corporate policy retrieval.

  • Key Metric: 96% Citation Accuracy, 93.5% Groundedness.
  • The Engineering: "Guardrail Prompt Template" + local CPU-based HuggingFace embeddings for cost-effective inference.
  • Design: Editorial/Brutalist UI, Vanilla HTML/CSS.

View Repository | View Live App


🎓 Education & Research

  • MSc, Software Engineering — Quantic (2025 - 2026)
  • MPhil, Intelligent Computing Systems — Ashesi University (2025–2027), research phase
    • Low-resource AI: lightweight adaptation, semi-supervised ASR, cross-lingual transfer for underrepresented languages
    • Security-first AI: responsible AI and accessible AI are the same problem

📊 Technical Arsenal

  • Languages: Python (Expert), TypeScript, Node.js, SQL.
  • Frontend: React, Next.js, Angular, Tailwind CSS.
  • Backend/DevOps: FastAPI, Django, Express.js, Docker, Kubernetes, RabbitMQ, AWS.
  • AI/ML: LangChain, PyTorch, HuggingFace, ChromaDB, AWS Bedrock.
  • Security: Ethical Hacking, OWASP Top 10 for LLMs.

📫 Connect

Pinned Loading

  1. x-shooter x-shooter Public

    Action-packed shooter game where players navigate through levels, defeat enemies, and aim for the highest score.

    JavaScript

  2. temperature-predictor temperature-predictor Public

    This project predicts London’s daily temperature using machine learning. It follows a structured ML workflow with EDA, data cleaning, model training, and experiment tracking with MLflow.

    Jupyter Notebook

  3. fusse-api fusse-api Public

    A Flask-based REST API backend for the Café Fausse restaurant website, providing reservation management, menu data, and newsletter subscription functionality.

    Python

  4. flower-classifier flower-classifier Public

    A command-line image classification application using PyTorch and transfer learning to identify 102 different flower species with custom training and prediction capabilities.

    HTML

  5. cafe-fusse-client cafe-fusse-client Public

    A modern, responsive restaurant website for Café Fausse, specializing in fresh Asian-inspired poke bowls and beverages. Built with React and designed for an exceptional user experience.

    JavaScript

  6. zamsi-ajegetina/insight zamsi-ajegetina/insight Public

    his is the FastAPI and LangChain implementation of the RAG Policy Q&A Application. It allows employees to ask questions about company policies and get grounded, cited answers generated by an LLM (L…

    Python