Jonathan Muk

Hi, I’m Jonathan Mukhobe

AI and ML Engineer

I build systems where a language model is one component among several: LangGraph agents with tool use and memory, retrieval pipelines on Qdrant, protocol-based tooling on MCP, and the FastAPI and React applications that put them in front of people.

Most of what I know is in the case studies. Kylie started as a WhatsApp agent with hand-wired tools and became an MCP-based system whose capabilities are discovered at runtime and traced end to end in Opik; Qdrant featured the original build. The Visual Document Understanding Pipeline turns PDFs and photos into clean Markdown: a layout model finds each region, Qwen3.5-4B on vLLM reads it, the GPU pools scale to zero, and an MCP server lets Claude read documents too.

Get in touchSee the case studies
Jonathan Mukhobe, AI and ML Engineer

What I do best

  • Automation

    Workflow automation that takes repetitive work off people: agents and pipelines that run on events or schedules, move data between the tools a team already uses, and recover from failures on their own, with queues, retries and monitoring built in.

    • Events and schedules
    • Queues
    • Retries
    • Monitoring

  • AI/ML Engineering & AI Agents

    Designing and deploying production-grade ML pipelines, autonomous AI agents, and LLM-powered systems. From LangGraph orchestration to vector memory and multi-modal processing, built for scale and reliability.

    • ML pipelines
    • LangGraph
    • Vector memory
    • Multi-modal

  • LLM Fine-tuning

    Adapting foundation models to specialised domains using supervised fine-tuning, RLHF, and parameter-efficient techniques (LoRA / QLoRA). Resulting in sharper, task-specific models that outperform general-purpose prompting.

    • SFT
    • RLHF
    • LoRA
    • QLoRA

  • Full Stack Engineering

    End-to-end product development from database design through REST APIs to polished React interfaces, delivered with performance, accessibility, and CI/CD built in from day one.

    • Databases
    • REST APIs
    • React
    • CI/CD

My Tech Stack

  • Programming

    3 tools

    • Python
    • JavaScript
    • TypeScript

  • AI tools and frameworks

    6 tools

    • LangChain
    • LangGraph
    • CrewAI
    • LangSmith
    • MCP
    • vLLM

  • Databases

    4 tools

    • PostgreSQL
    • MySQL
    • Qdrant
    • Redis

  • Backend

    2 tools

    • Django
    • FastAPI

  • Frontend

    3 tools

    • React
    • TypeScript
    • Tailwind CSS

  • DevOps

    3 tools

    • GitHub Actions
    • Docker
    • Grafana

Co-founded

Company I co-founded

PureNest360 Technologies and Real Estate logo

Co-Founder and Chief Technology Officer

PureNest360 Technologies and Real Estate

I co-founded PureNest360 and lead platform architecture, AI systems design and long-term technical strategy. I architected the real estate platform for the East African market and built both the technology and real estate web applications.

  • Kampala, Uganda
  • Founded 2025

Experience

Where I have worked

  1. May 2026 to present Kampala, Uganda Hybrid

    AI Engineer, Bright Labs, Bright Illuminated Thought Services

    • Develop and maintain in-house web platforms with AI integration, including the company flagship websites, for performance, responsiveness and brand consistency.
    • Build and deliver custom websites and full-stack web and AI applications for external clients, from requirements and wireframing through deployment and post-launch support.
    • Architect frontend interfaces in React and TypeScript with REST API integration for data-driven client solutions.
    • Run code reviews, performance audits and browser and device compatibility testing across delivered products.

  2. Feb 2026 to present Kigali, Rwanda Remote

    Lead Software and AI Engineer, Avenir Governance Advisory

    • Technical lead for the Innovation and Technology pillar, architecting the AI solutions behind the advisory blueprint.
    • Design and deploy autonomous AI agents and custom LLM workflows for organisational efficiency and decision support.
    • Build retrieval augmented generation systems and vector databases that turn institutional knowledge into searchable intelligence.
    • Implement CI/CD pipelines and MLOps workflows so AI deployments are scalable, secure and maintainable.

  3. Aug 2025 to present Kampala, Uganda On-site

    Co-Founder and Chief Technology Officer, PureNest360 Tech and Real Estate Company

    • Lead the technical architecture of a full-stack real estate platform for the East African market.
    • Design backend systems with scalability and cross-platform integration in mind.
    • Set up DevOps workflows and CI/CD pipelines through GitHub Actions to automate deployments.
    • Work with stakeholders to align product features with what the market actually needs.

  4. Mar 2025 to Sept 2025 Kampala, Uganda On-site

    Intern AI Engineer, Otic Foundation

    • Led technical workshops across Ugandan universities, driving AI awareness and unblocking student developers on integration problems.
    • Built intelligent agents with LangGraph, LangChain and CrewAI, focused on workflow automation and state management, monitored with LangSmith.
    • Created multimodal conversational systems with long-term memory on Qdrant.

Education

  • BSc Artificial Intelligence and Machine LearningISBAT University, Kampala, Uganda

Where the work has landed

Featured by Qdrant on XQdrant found Kylie through a LinkedIn post and featured it on X and LinkedIn.

Kylie - A Multimodal WhatsApp AI Agent

We recently came across this project shared by Jonathan Mukhobe Sr. on LinkedIn and wanted to highlight it for the community.

Kylie is a fully functional WhatsApp AI agent built with a practical, end-to-end…

Qdrant (@qdrant_engine),

View the post on X

110+

GitHub stars on Kylie

Counted 18 September 2026

34

Forks

Counted 18 September 2026

10+

Production web and AI projects shipped

Where it started

LinkedIn,

My original LinkedIn post about Kylie, the one that caught Qdrant's attention and led to the feature on X and LinkedIn.

Read the LinkedIn post