Skip to content
View Jerry-Khobby's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report Jerry-Khobby

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
Jerry-Khobby/README.md

Hi, I'm Jeremiah Anku Coblah πŸ‘‹

Data Engineer & Backend Engineer Β· Accra, Ghana πŸ‡¬πŸ‡­

I build the systems that move, clean, store and serve data:
batch and streaming pipelines, data warehouses, and the APIs and dashboards on top of them.

LinkedIn Email


🧭 About me

I'm a software and data engineer from Ghana, and a Computer Science graduate of Kwame Nkrumah University of Science and Technology (KNUST). I have about 2 years of experience in data engineering and 3 years building backend systems in Python and Node.js, across healthcare, payments, e-commerce and internal enterprise tools.

My work sits where software engineering and data engineering meet. On the data side, I build ETL/ELT pipelines with Airflow and dbt, stream real-time events through Kafka and Spark, and model data in Snowflake and Redshift so reports run fast and the numbers can be trusted. On the backend side, I design REST APIs, payment flows, authentication and real-time features, and I can build the React or Next.js dashboard that sits on top when a feature needs it.

What I enjoy most is making systems reliable. A pipeline that quietly loads wrong data is worse than one that fails loudly, so I build in validation, quality checks and clear error handling from the start. I like finding out why something is slow or broken by measuring it, rather than guessing, and I try to leave code documented and tested so the next engineer can work with it easily.

πŸ› οΈ What I do

πŸ—„οΈ Data engineering

  • Design and build ETL/ELT pipelines with Python, Apache Airflow and dbt, including scheduling, dependencies and failure handling
  • Build real-time streaming pipelines with Apache Kafka and Spark Structured Streaming
  • Model data warehouses in Snowflake and Amazon Redshift using star schemas for fast analytics
  • Build lakehouses with Delta Lake on AWS
  • Enforce data quality: validation, null checks, deduplication, lineage and rejected-record logging
  • Clean and transform data with Pandas and PySpark
  • Build web scraping workflows to automate data collection
  • Create reports and dashboards in Power BI, Grafana and Metabase

βš™οΈ Backend engineering

  • Design REST APIs with FastAPI, Django REST Framework, NestJS and Express
  • Integrate third-party APIs: Paystack payments, Cloudinary, SendGrid
  • Handle payment flows including escrow, failures and refunds
  • Build real-time features with WebSockets (Socket.IO)
  • Implement authentication and authorization: JWT, OAuth 2.0 (Google), role-based access control
  • Speed up apps with Redis caching and query optimization
  • Write tests with Jest and pytest, and document APIs with Swagger/OpenAPI
  • Build frontends with React, Next.js and Tailwind, and mobile apps with React Native

πŸš€ Featured projects

πŸ—„οΈ Data engineering

🎡 Music Streaming Data Pipeline An event-driven, serverless pipeline on AWS that ingests simulated music-streaming events and turns them into genre-level listening metrics. Events land in S3 and trigger EventBridge, are queued in SQS so bursts don't overwhelm processing, and are processed by Glue jobs (PySpark and Python Shell) orchestrated with Step Functions. CloudWatch logs every run and sends Slack alerts when a job fails, and the whole infrastructure is defined in Terraform so it can be rebuilt from scratch. S3 EventBridge SQS Step Functions Glue PySpark CloudWatch Terraform

πŸ›’ Lakehouse for E-commerce Transactions A small data lakehouse on AWS that loads e-commerce transaction data from S3 into Delta Lake tables. Before loading, it runs data quality checks (null checks, timestamp validation and deduplication) and logs rejected records for review instead of silently dropping them. The pipeline is automated with Step Functions, with CI/CD for the Glue scripts through GitHub Actions. AWS Glue PySpark Delta Lake Step Functions GitHub Actions

❀️ Real-Time Heartbeat Monitoring System A streaming system that simulates patient heart-rate data and processes it in real time. Events flow through Kafka into a Spark Structured Streaming job that validates the incoming JSON and stores clean records in PostgreSQL, where Grafana dashboards show the readings live. The full stack runs locally with Docker Compose. Kafka Spark Structured Streaming PostgreSQL Grafana Docker Compose

🏒 Office Management System (work project) The data and backend layer of a multi-tenant office management system covering 10+ business domains. I designed the relational schema, built an ETL pipeline that turns raw badge-reader events into daily attendance and punctuality records, and wrote 50+ SQL views providing attendance, occupancy and absenteeism KPIs for employees, managers, HR and Facilities teams. PostgreSQL SQL Views ETL REST APIs RBAC

βš™οΈ Backend & full-stack

πŸ›οΈ CampusBid: Campus E-commerce Platform A marketplace where students buy and sell safely, with payment held until the buyer confirms delivery. The NestJS backend has 10+ feature modules, an escrow-style Paystack payment flow that handles failures and refunds, real-time buyer-seller chat with Socket.IO, image uploads through Cloudinary, Redis caching and login with JWT or Google. It's covered by 167 unit tests across 14 suites and documented with Swagger/OpenAPI. NestJS MongoDB Redis Socket.IO Paystack Next.js Jest

🩺 Sepsis Prediction System My final-year project at KNUST: an end-to-end system that predicts sepsis risk in hospital patients from their vital signs and lab results over time. A hybrid CNN-LSTM model spots both sudden changes and gradual decline, a FastAPI backend serves single and batch predictions, and SHAP explanations show which factors drove each risk score so clinicians aren't looking at a black box. A Next.js dashboard supports CSV upload, real-time risk scoring and exportable PDF/CSV reports. TensorFlow Keras SHAP FastAPI SQLAlchemy Next.js

🎨 Real-Time Collaborative Whiteboard A drawing app where several people work on the same canvas at once and see each other's strokes instantly. Each canvas is its own room, so updates only reach the people on it, and drawing events are throttled to keep the app smooth under load. Drawings are saved in MongoDB so they're still there after a refresh. NestJS Socket.IO MongoDB Next.js Tailwind CSS


πŸ’Ό Experience

Data Engineer Β· AmaliTech Β· Nov 2025 – Oct 2026 Β· Kumasi

  • Built ETL/ELT pipelines with Python, Apache Airflow and dbt, cutting pipeline failure rates by an estimated 35%
  • Built and maintained warehouse models in Amazon Redshift and Snowflake, improving query performance by an estimated 45%
  • Built real-time streaming pipelines with Kafka and Spark, reducing data delivery lag by an estimated 40%
  • Enforced data quality and lineage checks across pipelines and warehouse layers
  • Built backend services and 50+ REST endpoints with role-based access for HR and Facilities dashboards

Backend / Data Engineer Β· BenchFive (Internship) Β· Sep 2024 – Jan 2025 Β· Remote

  • Built web scraping and ingestion workflows that automated previously manual data collection
  • Cleaned, transformed and validated datasets with Pandas before storage and analysis
  • Developed REST APIs with Django REST Framework, working with a Next.js frontend team

Agile Developer Β· Xcelsz (Contract) Β· Remote

  • Designed RESTful API endpoints for an internal business application within an Agile team

Backend Developer Β· Ideation Axis Group (Internship) Β· Sep – Nov 2024 Β· Remote

Mobile Application Developer Β· Zap Technology (Internship) Β· Oct – Dec 2023

  • Built mobile features with React Native, Redux Toolkit and Firebase

Web Developer Β· GI-KACE (Internship) Β· Sep – Oct 2023 Β· Accra

  • Built full-stack features with FastAPI, PostgreSQL, HTML, CSS and JavaScript

🧰 Tools & technologies

Languages

Languages

SQL

Backend frameworks

Backend

Django REST Framework SQLAlchemy Mongoose Socket.IO JWT OAuth 2.0

Frontend & mobile

Frontend

React Native

Databases

Databases

DynamoDB

Data engineering & big data

Kafka

Apache Airflow Apache Spark PySpark dbt Snowflake Amazon Redshift Delta Lake Pandas Selenium

Cloud & AWS

AWS

S3 Glue Lambda Step Functions EventBridge SQS CloudWatch

DevOps & tooling

DevOps

Jest pytest Swagger

Integrations

Paystack Cloudinary SendGrid

AI / ML

TensorFlow Keras SHAP Jupyter Matplotlib

Visualisation & BI

Power BI Grafana Metabase


πŸŽ“ Education & certifications

BSc Computer Science Β· Kwame Nkrumah University of Science and Technology (KNUST) Β· 2021 – 2025 Final-year project: an end-to-end sepsis prediction system using a CNN-LSTM model on clinical time-series data, with SHAP explainability.

AWS DataCamp DataCamp


πŸ“ˆ GitHub activity

GitHub streak

GitHub stats Top languages

Contribution graph


🀝 Let's work together

I'm open to data engineering and backend engineering roles, remote or any where in Ghana
The quickest way to reach me is by email or on LinkedIn.

Pinned Loading

  1. ghana-bus-booking ghana-bus-booking Public

    Book intercity bus seats in Ghana online: see which bus is loading, pick your seat on a live map, pay with mobile money and board with a QR ticket. Next.js, NestJS, PostgreSQL, Socket.IO.

    TypeScript 1

  2. Lakehouse-Architecture-for-E-Commerce-Transactions- Lakehouse-Architecture-for-E-Commerce-Transactions- Public

    A production-grade Lakehouse for e-commerce transactions. Raw CSVs land in S3, AWS Glue + Spark clean, validate, deduplicate and MERGE them into ACID Delta Lake tables, the Glue Data Catalog is ref…

    Python

  3. SikaTrack SikaTrack Public

    A personal-finance tracker for mobile-money users in Ghana. ("Sika" is Twi for money.) Most day-to-day money in Ghana moves through mobile money, which budgeting apps like Mint or YNAB can't see. S…

    TypeScript 1

  4. Music-Streaming-Data-Pipeline Music-Streaming-Data-Pipeline Public

    A production-grade, event-driven ETL pipeline built entirely on AWS that ingests music streaming events, transforms them through a three-tier data lake, computes daily KPIs per genre, loads the res…

    Python 2

  5. olist-dbt-analytics olist-dbt-analytics Public

    Analytics engineering project that transforms 100k real Brazilian e-commerce orders (Olist) into tested, documented dbt models on PostgreSQL. Covers staging, intermediate and mart layers, star sche…

    Python 1