Skip to main content

Senior Python / AI Platform Engineer

Technology
Pepla
Pretoria, South Africa1 month ago
Full timeOn-site

Job description

Company Description:

Pepla is a leading software development company based in Pretoria, South Africa, focused on delivering tailored solutions that optimize business operations. The team leverages deep technical expertise to create custom software, web applications, mobile apps, and enterprise systems that improve efficiency, control, and productivity. Pepla’s mission is to empower organizations with innovative, reliable technology that streamlines processes and frees them to focus on core priorities.

Applicants join a skilled, collaborative group dedicated to building cutting-edge solutions aligned with each client’s unique needs. Pepla positions itself as a long-term partner in progress for businesses seeking digital transformation.

Role Description:

We're hiring a Senior Python / AI Platform Engineer. This is a full-time, on-site role based in Pretoria for a US-based FinTech client building a next-generation mortgage platform for the American market. This role owns the production AI infrastructure — deploying, serving, and operating open-weight LLMs reliably at scale — while the client's ML Lead retains model strategy and methodology.

You'll be the engineer who turns AI capability into dependable, observable, production systems, integrating models into enterprise applications built on Python, FastAPI, Kafka, and MongoDB.

𝐓𝐡𝐢𝐬 𝐢𝐬 𝐚 𝐜𝐨𝐧𝐭𝐫𝐚𝐜𝐭-𝐭𝐨-𝐡𝐢𝐫𝐞 𝐞𝐧𝐠𝐚𝐠𝐞𝐦𝐞𝐧𝐭: 𝐚 𝐩𝐚𝐢𝐝 𝟗𝟎–𝟏𝟐𝟎 𝐝𝐚𝐲 𝐭𝐫𝐢𝐚𝐥, 𝐜𝐨𝐧𝐯𝐞𝐫𝐭𝐢𝐧𝐠 𝐭𝐨 𝐚 𝟏𝟐-𝐦𝐨𝐧𝐭𝐡 𝐬𝐭𝐚𝐟𝐟 𝐚𝐮𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 𝐜𝐨𝐧𝐭𝐫𝐚𝐜𝐭 𝐨𝐧 𝐬𝐮𝐜𝐜𝐞𝐬𝐬𝐟𝐮𝐥 𝐜𝐨𝐦𝐩𝐥𝐞𝐭𝐢𝐨𝐧. The client already has team leads and architects in place — you'll join a working engineering bench on a live project, not build a team from scratch.

Qualifications

  • 7+ years of professional software-engineering experience with strong production Python
  • Python production API/service development; FastAPI or comparable frameworks
  • Microservices, Kafka/event-driven systems, and MongoDB
  • Microsoft Azure and containers
  • Open-weight/local LLM deployment — Gemma or comparable models
  • Model-serving/inference APIs and structured model outputs
  • Model versioning, LLM evaluation/testing, and GPU/CPU inference environments
  • Observability, performance, and resource optimization
  • Experience with private/customer-controlled AI deployments
Keywords
FastAPIApplication Programming Interfaces (API)Apache KafkaScratchStructured FinanceEvent-drivenResource OptimizationOptimizationEngineeringLarge Language Models (LLM)TeaOCamlMongodbDevOpsPythonApache LicenseApache Http Server

Interested in this role?