Junior AI Platform Engineer

Junior AI Platform Engineer

Python and LLM Integration



Role overview

We are looking for a Junior AI Platform Engineer to help build and maintain AI features in Barq, an enterprise platform for analytics and customer-care applications. You will work with senior engineers on Python services, language-model integration, data preparation, testing, and deployment support.

This role suits an early-career engineer with solid programming fundamentals and evidence of practical learning through projects, internships, or professional work. You will own well-defined tasks with code review and mentoring, while developing experience in model evaluation and reliable AI applications.

Key responsibilities

• Implement and maintain small Python backend features, API integrations, and utility scripts within established designs.
• Help connect local and cloud language models to application workflows, including structured responses and tool calls, using existing interfaces.
• Prepare and validate training or evaluation datasets, checking formats, missing values, duplicates, and documented data provenance.
• Write and maintain unit and integration tests; reproduce bugs and verify fixes under senior review.
• Run established model evaluations and bounded performance tests, recording inputs, settings, results, failures, and resource measurements.
• Investigate application logs and request traces, document findings, and escalate unclear or cross-component issues with reproducible examples.
• Assist with Docker-based development, CI checks, and deployment verification using approved runbooks and supervision.
• Follow existing authorization, data-handling, and secret-management practices; keep sensitive information out of logs and shared artifacts.
• Update setup guides, test instructions, and change notes, and participate in code reviews and team discussions.

Scope and support

Senior engineers will guide architecture, security-sensitive changes, production releases, fine-tuning strategy, and GPU resource configuration. You will contribute through scoped implementation, evidence collection, testing, and documented experiments, taking on broader responsibility as your skills develop.

Required qualifications

• Working knowledge of Python, including functions, classes, exceptions, modules, and common data structures.
• Understanding of HTTP APIs, JSON, and client-server request flows, with experience building or consuming an API in a project.
• Basic SQL skills and familiarity with relational databases and structured data.
• Ability to use Git for branches, commits, and pull requests, and to work from the command line.
• Familiarity with automated testing and a systematic approach to reproducing and debugging problems.
• Basic understanding of LLM prompts, context, structured outputs, and why generated answers and code require validation.
• Ability to explain your own code, respond to review feedback, and communicate progress and blockers clearly.
• Practical evidence of programming ability through personal, academic, open-source, internship, or professional projects. Previous production AI experience is not required.

Preferred qualifications

• Exposure to FastAPI or another Python web framework, PostgreSQL, and Docker.
• A project involving an LLM API, embeddings, retrieval, or a locally hosted model.
• Familiarity with pytest, GitHub Actions, Linux, or basic asynchronous programming.
• Experience cleaning or transforming data with pandas, SQL, CSV, JSON, or Parquet.
• Exposure to React or TypeScript, Arabic-English AI workflows, or introductory machine-learning coursework.

Learning opportunities

You will build experience in reliable tool calling, model evaluation, observability, and concurrent application workflows. Depending on project needs, you may assist with supervised fine-tuning experiments, GPU inference, and larger data-processing systems. Prior expertise in these areas is not required.

Expected outcomes

• Deliver scoped features and bug fixes that pass relevant checks and senior code review.
• Produce reproducible test and evaluation results, with failures and limitations clearly recorded.
• Follow data-access and operational procedures and raise risks or blockers promptly.
• Maintain useful documentation and gradually handle routine debugging and implementation with less supervision.