Skip to content
View Shotza247's full-sized avatar

Block or report Shotza247

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
Shotza247/README.md

Hi, I'm Jabulani Ndlovu👋

Early-Career Enterprise Data & ML/AI Engineer | Agentic AI & Analytics Engineering

I build data-driven and AI-powered solutions that transform complex data into actionable insights. My experience spans enterprise analytics, data engineering, intelligent automation, AI agents, API integrations, and scalable data platforms. I am passionate about building trustworthy AI systems powered by strong data foundations.


🎯 My Objective

I am an early-career engineer working at the intersection of enterprise data platforms and applied ML/AI. My goal is to grow into an engineer who can take a business problem from raw data to a governed, production-ready AI solution:

  • Data foundations first — ingesting, modelling, governing and serving trusted data (batch and real-time) that AI systems can rely on.
  • Applied ML/AI that ships — moving beyond notebooks to full-stack apps, APIs, RAG pipelines, agents and MLOps/LLMOps practices.
  • Enterprise-grade thinking — security, PII handling, evaluation, observability and governance are part of the design, not an afterthought.
  • Business-value communication — documenting clearly and explaining technical work to stakeholders in terms of outcomes.

This profile is my public learning log: every repository below is a step towards that goal, from first Python scripts to agentic, retrieval-augmented and real-time data systems.


💼 Experience

IBM — Data & ML/AI Engineering Intern

  • Building enterprise data pipelines using IBM Watsonx.data, AI-enabled analytics workflows using IBM watsonx Orchestrate and watsonx.ai, intelligent automation solutions, and agentic systems that help demonstrate how data can be prepared, governed, integrated, and automated to support business outcomes.
  • Using IBM Bob as an agentic coding partner to accelerate delivery, e.g. the Pulse360 HR analytics hackathon build.
  • Prototyping MCP-based tool integrations so agents can connect to enterprise data and APIs through a reusable, governed pattern.

🔧 Currently Working On

  • 📚 Back-End AI Engineering Intern — Starting 1 July with FlyRank.ai -- Objective : Full End-to-End Capstone Project

  • 📊 Data Analytics & Engineering — Local Data pipelines, Databricks pipelines, and building small ML/AI-enabled insights streamlit powered apps

  • 📈 Self-taught python — navigating the advanced features and libraries within python to elevate my current skill-set

  • 🌐 Personal brand — improving my writing (Documentation), speaking (Stakeholder/Business-value communication)

  • TODO: Building in public (Coming Soon...)


🚀 Current Career Focus

  • Agentic AI Systems
  • Analytics Engineering
  • Data Engineering
  • Data Science & Machine Learning
  • Intelligent Automation
  • AI Evaluation & Governance

🚀 Featured Projects

Project What it demonstrates Stack
Milvus_4RAG_Template Production-ready, modular RAG template: Milvus vector DB, 2-stage cross-encoder reranking, LLM synthesis, synthetic benchmarking and real-time RAG Triad evaluation in a Streamlit chat studio. Podman & Docker ready. Milvus, FastAPI, LangChain, Sentence-Transformers, RAGAS, Streamlit
OpenSource-RAG-Template Open-source FAQ RAG agent with microservices that search and synthesize retrieved chunks. Qdrant, FastAPI, Streamlit
Pulse360_Hackathon_Day Full-stack internal HR analytics platform for monthly/quarterly review cycles: employee contributions, hiring conversions and system adoption. Built with IBM Bob. TypeScript, Supabase/PostgreSQL, MCP, OpenAI
Employee_Full-Stack_Churn_ML End-to-end churn prediction app: CSV import, missing-value handling, model training and prediction exploration. Streamlit, FastAPI, SQLite, scikit-learn
Data-Analytics-AI-Assistant Streamlit app for exploring CSVs with NLP, automatic summaries, PII/SPI detection, AI-generated analysis and PDF output. Streamlit, Pandas, OpenAI, Matplotlib
IBM-WxO_ADK_SetUp watsonx Orchestrate agent backed by an MCP server via the ADK, using the M+N integration pattern (agent → toolkit → tools → API). watsonx Orchestrate, FastMCP, Open-Meteo API
File_Download_MCPServer_Tool Reusable local MCP server with GPT-4o tool calling and local file writing. MCP, OpenAI, Python
GraphQL-Server-SetUp.Net GraphQL API used as an agent tool in an employee retention/promotion POC & MVP. C#, .NET, GraphQL, MongoDB Atlas
Data-Streaming Real-time streaming analytics: consumer-group based ingestion with schema registry feeding a live dashboard (see below). Python, Streamlit, Schema Registry

📸 Project Snapshots

Real-Time Tax Applications Analytics Dashboard — live streaming consumer with auto-refresh, buffered topics (application_analytics, customer_risk_profile), schema registry support and filterable views by status, tax year, employment type and province.



Agent Tool-Calling Sequence (watsonx Orchestrate + MCP) — a user question flows through the agent, toolkit and MCP server to an external API and back as a natural-language answer.

🧪 Machine Learning Foundations


🧭 How I Work

  1. Understand the business question before choosing tools.
  2. Prepare and validate the data (quality checks, PII awareness, schemas).
  3. Build small, working end-to-end slices (data → model/agent → API → UI).
  4. Evaluate honestly (metrics, RAG Triad, benchmarks) and iterate.
  5. Document and communicate so others can reuse and trust the work.

🛠️ Technical Skills

Languages

  • Heavily Used Day-to-Day work:

Python SQL

  • Heavily Used for API's:

JavaScript C# HTML5 CSS3

Data & Analytics

  • Heavy Interaction:

Excel Pandas NumPy Matplotlib Seaborn Scikit-Learn

  • Foundational Exploration:

Power BI

Data Engineering

DB2 Watsonx.data Apache Spark Databricks Presto SQL ETL Pipelines REST APIs GraphQL MongoDB

AI & Machine Learning

Watsonx.ai Azure ML Studio LLMOps Prompt Engineering AI Agents MLOps Model Evaluation

Frameworks & Platforms

.NET Android Studio Azure Azure ML Studio Databricks Watsonx.data Watsonx Orchestrate SQL Server MongoDB Supabase

Tools & IDEs

Visual Studio Git GitHub VS Code Jupyter Docker Postman


🌱 Currently Learning

Technology Focus IBM Tools
Python Python Data Engineering/Science/Analytics & Automation watsonx.ai L3
Azure Azure Fundamentals AZ-900 Certification watsonx.data L4
Azure ML Studio Azure ML Studio Up-skill on Azure Data Science Solutions CP4D/ IBM Data Fabric / MDM
Databricks DataBricks Platform Playground watsonx Orchestrate
MongoDB MongoDB NoSQL DB for LLM/NLP/AI-infused Projects watsonx.data Intelligence/Integration L3
Supabase Supabase RDB (postgres arch) for Projects IBM BOB L3

📊 GitHub Stats

GitHub Streak

Top Languages My GitHub Stats


📡 How to Reach Me

LinkedIn Dev.to


Building intelligent systems starts with building trusted data foundations.

Pinned Loading

  1. California_Housing_PredictionV1 California_Housing_PredictionV1 Public

    Predicts California district-level house prices using census data — combining KNN imputation, feature engineering, and a Random Forest model (R² = 0.75) to help property businesses make faster, dat…

    Jupyter Notebook

  2. Chronic_Kidney_Disease_ClassV1 Chronic_Kidney_Disease_ClassV1 Public

    A machine learning project using anonymized blood sample data to classify and predict Chronic Kidney Disease across different age groups and medical aid membership bands.

    HTML

  3. GraphQL-Server-SetUp.Net GraphQL-Server-SetUp.Net Public

    This GraphQL project was part of an Employee Retention/Promotion POC & MVP, that would modernize HR processes in the Agentic age to be able to retain their Talent and also promote them incase someo…

    C#

  4. IBM-WxO_ADK_SetUp IBM-WxO_ADK_SetUp Public

    A beginner project demonstrating how to build a watsonx Orchestrate agent backed by an MCP server using the ADK. Covers the M+N integration pattern — agents connect to one MCP toolkit, the toolkit …

    Python