Software Developer, Software Engineer, AI Engineer

Overview

We are looking for a Software Developer for one of our clients with the following details:

Location:
Abu Dhabi, UAE

Start Date: ASAP

Contract Duration: 1 year

Education / Experience

• Bachelor's degree in Computer Science, Engineering, or equivalent practical experience.
• 5+ years of professional software development experience.
• Demonstrated project delivery in Python (portfolio, Git Hub, internal repos, or verifiable shipped work).

Required Qualifications

• Strong Python development skills, demonstrated through delivered projects in production or near production environments (APIs/services, data workflows, automation, or platforms).
• Experience building software with clear engineering fundamentals: modular design, testing, version control, packaging, performance considerations.
• Ability to collaborate effectively with domain SMEs and data scientists, can translate ambiguous domain goals into concrete, testable software.
• Experience working with data-centric systems (structured/unstructured data handling, ETL/ELT concepts, data quality, metadata).
• Familiarity with modern development workflows, specifically with Azure Dev Ops (ADO) for Agile work management (Azure Boards), source control (Azure Repos), and/or CI/CD pipelines (Azure Pipelines).

Preferred Qualifications

• Familiarity with GenAI concepts and building blocks: LLM APIs, RAG pipelines, embeddings/vector databases, prompt design, and evaluation.
• Experience with agent frameworks and orchestration patterns (e.g., Lang Graph / Semantic Kernel / similar), and tool-calling architectures.
• Exposure to "agentic workflow" patterns (goal-oriented loops, plan/act/observe/revise, human-in-loop gating).
• Experience deploying containerized services (Docker/Kubernetes) and building runtime-ready APIs. Understanding of petrotechnical workflows and data types (subsurface interpretation, logs, reservoir/simulation workflows).

• Builder mindset: ships working software, iterates fast, improves reliability over time.
• Strong written and verbal communication: can explain tradeoffs to SMEs and engineers.
• Pragmatic problem solver: balances GenAI "flexibility" with engineering rigor and validation.
• High ownership and integrity around data handling, access control, and compliance expectations in controlled environments

Build Agentic Workflow Software (Core)

• Implement agentic workflows that execute multi-step domain tasks with human-in-the-loop checkpoints (review/approve/iterate) and clear auditability.
• Develop modular, extensible components (agents, tools, planners, evaluators) so workflows can be reused and expanded across use cases.
• Translate SME requirements into software artifacts: workflow definitions, task decomposition, tool contracts, state management, and error handling.

Integrate with Data & Domain Ecosystems

• Build robust integrations to enterprise data foundations (e.g., OSDU/ADME) for reading inputs and writing outputs with metadata/provenance.
• Implement connectors to domain tooling and services (e.g., subsurface/petrotechnical apps and workflows), supporting orchestration across systems.
• Collaborate with data engineers/data scientists to operationalize pipelines that support workflow execution and repeatability.

GenAI Enablement (Preferred)

• Implement LLM-enabled features: prompt/tool orchestration, retrieval-augmented generation (RAG), document ingestion, embeddings, and vector search.
• Contribute to multi-agent workflow frameworks where LLMs assist with reasoning/planning and tools execute deterministic steps.
• Apply best practices for reliability (guardrails, evaluation, regression tests, and monitoring) in GenAI systems.

Engineering Excellence & Operational Readiness

• Write clean, testable Python code with CI/CD, code review, and documentation discipline.
• Add observability: logging, metrics, traces, structured events, and workflow run histories (supporting auditability).
• Engineer workflows for production standards (security, performance, resiliency), with a bias toward maintainable systems (not "one-off demos").

If Interested Kindly apply.