Soum

Soum

Soum is a fast-growing recommerce marketplace redefining buying and selling of used electronic devices in MENA with trust, security, and money-back guarantee.

Household Durables
11-50

Description

  • Architect production GenAI systems across conversational agents, workflow automation, personalized discovery, search relevance, content generation, and trust-and-safety use cases.
  • Own features end-to-end from problem framing and model selection through backend implementation, frontend integration, deployment, evaluation, and monitoring.
  • Design agentic workflows with tool calling, multi-step reasoning, retrieval-augmented generation, and integrations with internal APIs and external SaaS tools.
  • Build and maintain the retrieval and embeddings stack, including chunking, embedding selection, vector indexing, hybrid search, reranking, and retrieval evaluation.
  • Improve reliability and cost efficiency through streaming, prompt caching, latency management, token optimization, observability, and fallback handling.
  • Establish offline and online evaluation frameworks for response quality, tool-call accuracy, hallucination rate, retrieval precision, and business impact.
  • Evaluate new models, frameworks, and agent patterns, run experiments, and ship effective approaches into production.
  • Mentor engineers on AI/ML best practices, prompt engineering, and production readiness.
  • Partner with Product, Engineering, Data, Operations, and Customer Experience teams to identify and deliver high-leverage AI opportunities.

Requirements

  • 5+ years of production software engineering experience, including at least 2 years shipping LLM- and ML-powered features at scale.
  • Strong Python skills for backend services, scripting, data pipelines, and ML tooling.
  • Working ability with TypeScript and React for integrating AI features into product surfaces.
  • Production experience with major LLM providers such as Gemini, Claude, or GPT, including tool/function calling, structured outputs, streaming, prompt caching, and cost control.
  • Deep understanding of Retrieval Augmented Generation, including document ingestion, chunking, embeddings, vector databases, hybrid retrieval, and reranking.
  • Solid knowledge of embeddings and vector search, including dense and sparse representations, similarity metrics, HNSW/IVF indexing, and dimensionality tradeoffs.
  • Strong ML and NLP fundamentals, including transformer architectures, tokenization, fine-tuning vs. prompting, classification, ranking, and evaluation.
  • Experience building scripting and automation for data preparation, model evaluation harnesses, and offline analysis.
  • Comfort working with PostgreSQL, Redis, REST, SSE, WebSockets, and event-driven architectures.
  • Production experience with observability, A/B testing, and safe rollout of model changes.
  • Experience with recommendation systems, learning to rank, or search relevance at scale is preferred.
  • Marketplace or C2C experience covering buyer/seller dynamics, disputes, fraud, and payouts is preferred.
  • Multimodal model experience, including vision and image understanding for listings, is preferred.
  • Arabic NLP or bilingual product experience is preferred.
  • Experience with agent frameworks such as LangGraph or custom orchestrators is preferred.
  • Experience with fine-tuning, LoRA, distillation, or hosting open-weight models in production is preferred.
  • Open source contributions to LLM tooling, eval frameworks, or retrieval libraries are preferred.

Interested in this position?

Apply directly on the company website

Apply Now

Similar Roles

AI Evaluation & Annotation Reviewer (L3 - Advanced Level) - French

Volga Partners 51-250 Internet Software & Services

AI Evaluation & Annotation Reviewer at an artificial intelligence project, evaluating and annotating French language content to improve the quality and reliability of AI-generated responses.

Machine Learning
11 hours, 23 minutes ago

AI/ML Developer (LLMs, NLP, and AI Applications)

Pakistan Single Window 51-250 Internet Software & Services

An AI/ML Engineer at the company will develop and deploy LLM and NLP applications for intelligent question answering, chatbots, and real-time production use.

Django Docker FastAPI Flask Kubernetes Python PyTorch SageMaker TensorFlow
11 hours, 23 minutes ago

Principal AI & Cloud Security Engineer (Remote only within WA, ID, OR)

Banner Bank 1K-5K Banks

Banner Bank is seeking a Principal AI & Cloud Security Engineer to define enterprise security strategy for AI-powered systems and Azure environments across the organization.

Azure Cybersecurity DevSecOps Go LLM Python Rust Terraform
11 hours, 23 minutes ago

LLM QA Contractor (Part-Time | Remote | Freelance) | Image & Text Review

Alongside Internet Software & Services

Alongside is hiring a detail-oriented LLM QA Contractor for an AI validation project that reviews image- and instruction-based outputs to improve the accuracy and reliability of LLM responses.

LLM Python
11 hours, 23 minutes ago

You're on a roll! Sign up now to keep applying.

Sign Up

Already have an account? Log in

Used by 14,729+ remote workers