Reports to: Product / Technical Lead
As an AI & Machine Learning Engineer at ERGO, you will design and build intelligent product capabilities using generative AI, machine learning, and data-driven systems.
You will work across product, design, and engineering teams to turn complex requirements into reliable AI-powered features — from retrieval and structured generation to intelligent decision-support systems. Your focus will be on building AI that works in real products: grounded, testable, measurable, and robust enough to move beyond the prototype stage.
This role suits someone who is comfortable moving between experimentation and engineering, and who understands how to turn modern AI capabilities into dependable product behavior.
Build AI-Powered Product Capabilities
- Design and implement AI and machine-learning features that support real product workflows.
- Integrate large language models, machine-learning models, and external AI services into digital products.
- Build multi-step AI workflows involving retrieval, reasoning, validation, tool use, and human review.
- Translate product requirements into technical AI behavior, model inputs, outputs, and decision logic.
- Work with software engineers to expose AI capabilities through reliable services and APIs.
Develop Generative AI & Retrieval Systems
- Build retrieval-augmented generation pipelines using structured and unstructured source material.
- Design document ingestion, chunking, indexing, retrieval, and context-selection workflows.
- Implement structured generation using schemas, validation, and deterministic constraints where appropriate.
- Improve grounding, consistency, and traceability of generated outputs.
- Design fallback and failure behavior when the system cannot produce a sufficiently reliable result.
Evaluate AI & ML Systems
- Build evaluation datasets, test cases, and benchmarks for important system behaviors.
- Measure model quality, retrieval performance, consistency, accuracy, and failure modes.
- Compare models, prompts, retrieval strategies, and system architectures using repeatable evaluations.
- Investigate poor outputs and trace problems back to data, retrieval, prompts, model behavior, or system logic.
- Define technical acceptance criteria before AI capabilities are released into live workflows.
Prototype, Integrate & Harden
- Build technical prototypes to test high-risk assumptions before larger engineering investment.
- Move successful prototypes into maintainable production components.
- Document architecture, dependencies, model choices, limitations, and technical decisions.
- Work with product and design teams to test AI behavior within real user workflows.
- Identify what is ready for production and what still requires further hardening.
We are looking for someone who understands that good AI products require more than a model API or a promising demo.
You may come from software engineering, machine learning, applied AI, data science, computer science, or a related field.
You likely bring:
- 4–6 years of relevant experience in software engineering, machine learning, applied AI, or data-intensive product development.
- Strong Python development skills.
- Hands-on experience integrating LLMs into working applications.
- Practical experience with RAG, embeddings, vector search, and document-processing pipelines.
- Experience with prompt design, structured outputs, tool/function calling, and multi-step AI workflows.
- Experience building evaluation approaches for generative-AI or machine-learning systems.
- A practical understanding of hallucination, grounding, context limitations, and other common AI/ML failure modes.
- Experience working with APIs, databases, backend services, and production software systems.
- Familiarity with version control, testing, deployment, and modern software-development practices.
- The ability to explain technical trade-offs clearly to product and design teams.
Additional experience that would be valuable:
- Building predictive or personalization models for classification, ranking, recommendation, or prediction.
- Combining deterministic rules with machine-learning approaches in hybrid systems.
- Working with structured or behavioral data for feature development, experimentation, or model evaluation.
- Strong SQL skills and experience with data preparation, event models, or data pipelines.
- Multilingual AI or complex document-intelligence workflows.
We are particularly interested in people who can distinguish between an impressive prototype and a reliable, measurable product capability.
Please submit your CV, GitHub or technical portfolio, and a short introduction explaining why this role interests you and what you would bring to ERGO.
Include 2–3 relevant projects showing:
- The problem you were solving.
- Your technical contribution.
- How AI or machine learning was used.
- The system architecture or workflow you designed.
- How you evaluated model quality and handled failure modes.
- What you changed to improve reliability, performance, or user value.
Location: Cairo, Egypt
Deadline to apply: 01 October 2026
Start date: 1 November 2026
Contract type: Full-time
Work Model: 3-2-2 Hybrid