04 / Portfolio guide

Ask a focused question. Follow the source.

Verified mode searches a curated set of published portfolio answers. Optional Local AI can interpret a question on the visitor's device, but it only ranks those same inspectable answers.

Curated answersBrowser-only matchingOptional Local AISource links included

Ask about the work.

Verified mode matches your question to published portfolio facts in your browser. It does not generate new claims or send your question to an AI service.

Verified guide ready. Choose a suggested question or type your own.

No question selected

Choose what you want to know.

Start with a suggested question, or search the published portfolio in your own words.

Curated answer library

Published facts, linked to their source.

Full-stack range with back-end depth

Phadima is a software engineer with full-stack experience and strong back-end depth. He works across interfaces, services, data, cloud delivery, and operations, with particular attention to clear contracts and dependable behavior.

Profile and capabilities

Eight rules with an explanation for every decision

The Fraud Rule Engine combines transaction, recent-history, and customer-baseline signals in a .NET 10 service. Eight rules feed one scoring policy, every assessment records its rule outcomes and rule-set version, and the API returns Approve, Review, or Decline.

Fraud Rule Engine case study

Agent events become idempotent, queryable alerts

The Agent Activity Risk Monitor uses Fastify for ingestion and reads, Python for five pure analysis rules, and PostgreSQL as the shared contract. Unique events and alerts plus concurrent row claiming keep retries and multiple analyzers safe.

Agent Activity Risk Monitor case study

A product-expiry platform in development

ShelfIQ is a private-source .NET 8 product in development for capturing shelf-life data, classifying expiry status, verifying barcodes, summarizing inventory, and handling product images through a separate boundary. It has not been publicly launched, and its page lists the gates that remain before release.

ShelfIQ project story

Product, infrastructure, and financial systems

Phadima currently builds back-end services at Capitec Bank. Earlier roles include Amazon EC2, OLX, independent software development, and programming education. The journey route lists the dates, scope, and technologies for each role.

Career journey

Back-end systems and APIs

His published back-end toolkit includes C#, .NET, Java, Python, Node.js, REST APIs, SQL Server, PostgreSQL, DynamoDB, Redis, and Elasticsearch. His portfolio emphasizes explicit domain models, stable contracts, secure boundaries, testing, and operational evidence.

Back-end capabilities

Cloud delivery backed by operational work

Phadima has worked with AWS and Azure, Docker, CI/CD, observability, incident response, and DevSecOps. At Amazon EC2, he built orchestration and self-service APIs for fleet provisioning, maintenance, patching, and infrastructure lifecycle work.

Career and cloud capabilities

Data models, queries, and analysis

His published work covers SQL Server, PostgreSQL, MySQL, DynamoDB, MongoDB, Redis, Elasticsearch, data modelling, and query tuning. The SIC project also uses Python, Pandas, Jupyter, and visualization to examine unemployment, poverty, and crime in South Africa.

Work and data capabilities

Application development and software engineering

Phadima has studied application development, software engineering, ICT applications development, and Python programming. His profile shows eight recent and ten selected certifications from 114+ listed on LinkedIn.

Education and certifications

Six working languages

Phadima lists Sepedi and English at native or bilingual proficiency, Sesotho and Setswana at full professional proficiency, IsiXhosa at professional working proficiency, and IsiZulu at limited working proficiency.

Languages

Start with the contact form

Use the portfolio contact form for a role, project, collaboration, speaking, or mentoring conversation. The profile also links to Phadima on GitHub and LinkedIn.

Contact options