Only AI Jobs


AI Engineer | Ingénieur(e) en IA

ID: 9146

Type: Full-time

Category: Others

Company Name: O2 Commerce

Location: Iowa (USA)

Education Level: Senior (5-10 years)

Visit company vacancy
Job Description

(EN) Datmos brings together passionate experts who combine creativity and data to craft inspiring eCommerce experiences.

Driven by a shared passion for excellence, we help visionary brands grow by aligning commerce strategy, digital experience, and technology implementation to deliver innovative, data-driven digital commerce solutions.

(FR) Datmos réunit des experts passionnés qui combinent créativité et data pour imaginer des expériences e-commerce inspirantes.Portés par une même passion pour l’excellence, nous aidons les marques à croître en alignant stratégie commerciale, expérience numérique et mise en œuvre technologique, afin de livrer des solutions de commerce digital innovantes et centré sur la donnée.


_________________________________

(FRENCH WILL FOLLOW)


Datmos is hiring an AI Engineer !


Reporting to the Director of AI, the AI Engineer is responsible for the technical architecture of our intelligence systems, focusing on the design and implementation of agentic workflows and RAG systems. You translate the AI strategy into functional logic, ensuring that models produce precise, secure, and scalable results. As a technical lead, you set the standards for model selection and prompt engineering while mentoring others on the team. You collaborate closely with developers to integrate these "brains" into final products and stay at the cutting edge of model performance and optimization.


Key Responsibilities : 


  • Agentic System Design: Develop and optimize complex, multi-step agentic workflows that allow AI to perform autonomous tasks and decision-making.
  • RAG Architecture: Design and maintain Retrieval-Augmented Generation systems to connect LLMs with proprietary data securely and accurately.
  • Model Selection & Tuning: Evaluate and configure various LLMs (OpenAI, Claude, Llama) based on specific project constraints like latency, cost, and data privacy.
  • Advanced Prompt Engineering: Establish standards for instruction design to ensure model outputs are reliable, safe, and contextually aware.
  • Context & Memory Architecture: Design short-term and long-term memory strategies for agentic systems, including conversation summarization, vector store management, and session state handling.
  • Guardrails & Safety Systems: Implement input/output validation layers, content filtering, and fallback logic to ensure model outputs remain within acceptable boundaries for client-facing applications.
  • AI Observability & Logging: Instrument AI pipelines with tracing, logging, and monitoring tooling (e.g., LangSmith, Langfuse) to debug production failures and track model drift over time.
  • Technical Best Practices: Document workflows for AI logic and create reusable templates for prompt chains and model configurations.
  • Continuous Innovation: Experiment with new frameworks to improve the sophistication of our AI offerings.


Required Skills & Competencies : 


  • Agency Experience: 3+ years of experience in an agency environment (or fast-paced consulting).
  • AI Proficiency: Demonstrated interest in AI tools with at least 2-3 documented use cases applied to current or past projects.
  • Experience: 5+ years in software engineering, with strong coding skills (Python/Backend).
  • Technical Depth: Experience with Agents, MCP, or AI related software orchestration, Datalakes( BigQuery, Databricks, Snowflake )
  • LLM Proficiency: Advanced knowledge of prompt engineering techniques and the specific nuances/strengths of major foundational models.
  • Cloud : Development of cloud architectures that are optimized for real-time & AI
  • Orchestration Frameorks: Hands-on experience with agentic frameworks such as LangChain, LlamaIndex, CrewAI, AutoGen, or custom tool-calling implementations using native model APIs.
  • Vector Database Expertise: Practical experience with vector stores (Pinecone, Weaviate, pgvector, ChromaDB) including indexing strategies, similarity search tuning, and hybrid retrieval approaches.
  • Fine-Tuning & Adaptation: Familiarity with parameter-efficient fine-tuning methods (LoRA, QLoRA) and when to apply fine-tuning vs. prompt engineering vs. RAG for a given use case.
  • AI Security: Understanding of prompt injection, jailbreaking vectors, and data leakage risks specific to LLM applications — and the ability to design mitigations at the architecture level.
  • System Architecture: Ability to design scalable technical workflows that handle high-volume AI requests efficiently.
  • Problem-Solving: Strong ability to debug "black box" model behaviors and iterate until precision benchmarks are met.
  • Relevant certifications (an asset)


Joining Datmos is : 

Reaching your full potential : 

  • Challenge yourself: By working with major groups and well-known brands in North America (e.g., Group CH, Five Below, Skullcandy, Yeti Cycles, Casper, Jackson Mountain Resort, and many others), as well as with renowned digital partners such as Bloomreach, Shopify, Big Commerce, Akeneo, Klaviyo, Adobe, etc.
  • Contribute to the evolution of our vision: Bring your innovative ideas and expert perspective.
  • Learn from the best: Come work with over 100 certified experts in one of the most recognized agencies in the market.
  • Push yourself by developing your expertise: Access all the certifications you want and train while being paid.


Putting yourself first : 

  • Access to health Insurances paid in part by the employer (Group insurance for you and your family). 
  • Vacations and mobile days to take some rest. 
  • Disconnection policy to promote work-life balance 
  • Flexibility in your work location and your work hours (Offices in the US and Canada or possibility to work 100% remote)
  • Access to a recognition program in the form of gift cards or donations to associations. 


*Datmos has been recognized as one of the best workplaces in Canada by Great Place to Work since 2022.

Don’t wait any longer and join the team!


___________

FRANÇAIS:



Datmos recrute un.e Ingénieur(e) en IA 


Relevant du / de la Directeur(trice) IA, L’ingénieur(e) en IA est responsable de l'architecture technique de nos systèmes d'intelligence, se concentrant sur la conception et l'implémentation de flux de travail (workflows) agentiques et de systèmes RAG. Vous traduisez la stratégie IA en logique fonctionnelle, garantissant que les modèles produisent des résultats précis, sécurisés et évolutifs. En tant que leader technique, vous fixez les normes pour la sélection des modèles et le "prompt engineering" tout en agissant comme mentor pour les autres membres de l'équipe. Vous collaborez étroitement avec les développeurs pour intégrer ces « cerveaux » dans les produits finis et restez à la fine pointe de la performance et de l'optimisation des modèles.


Responsabilités clés : 


  • Conception de systèmes agentiques : Développer et optimiser des workflows agentiques complexes et multi-étapes permettant à l'IA d'exécuter des tâches et des prises de décision autonomes.
  • Architecture RAG : Concevoir et maintenir des systèmes de génération augmentée par récupération (RAG) pour connecter les LLM aux données propriétaires de manière sécurisée et précise.
  • Sélection et réglage de modèles : Évaluer et configurer divers LLM (OpenAI, Claude, Llama) en fonction des contraintes spécifiques aux projets, telles que la latence, le coût et la confidentialité des données.
  • Prompt Engineering avancé : Établir des normes de conception d'instructions pour garantir que les sorties des modèles sont fiables, sécurisées et adaptées au contexte.
  • Architecture de contexte et mémoire : Concevoir des stratégies de mémoire à court et long terme pour les systèmes agentiques (résumé de conversation, gestion des entrepôts vectoriels et gestion de l'état des sessions).
  • Garde-fous et systèmes de sécurité : Implémenter des couches de validation d'entrée/sortie, du filtrage de contenu et des logiques de repli (fallbacks) pour garantir que les réponses restent dans les limites acceptables pour les applications clients.
  • Observabilité et journalisation IA : Outiller les pipelines IA avec des outils de traçabilité et de surveillance (ex: LangSmith, Langfuse) pour déboguer les erreurs en production et suivre la dérive (drift) des modèles au fil du temps.
  • Meilleures pratiques techniques : Documenter les flux de travail pour la logique IA et créer des modèles réutilisables pour les chaînes de prompts et les configurations de modèles.
  • Innovation continue : Expérimenter avec de nouveaux frameworks pour améliorer la sophistication de nos solutions IA.



Compétences et qualifications requises : 


  • Expérience en agence : 3 ans et plus d'expérience dans un environnement d'agence (ou en consultation dans un environnement au rythme rapide).
  • Maîtrise de l'IA : Intérêt démontré pour les outils d'IA avec au moins 2-3 cas d'utilisation documentés et appliqués à des projets actuels ou passés.
  • Expérience : 5 ans et plus en génie logiciel, avec de solides compétences en programmation (Python/Backend).
  • Profondeur technique : Expérience avec les agents, le protocole MCP ou l'orchestration logicielle liée à l'IA, et les lacs de données (BigQuery, Databricks, Snowflake).
  • Maîtrise des LLM : Connaissance approfondie des techniques de prompt engineering et des nuances/forces spécifiques des principaux modèles fondateurs.
  • Cloud : Développement d'architectures cloud optimisées pour le temps réel et l'IA.
  • Frameworks d'orchestration : Expérience pratique avec des frameworks agentiques tels que LangChain, LlamaIndex, CrewAI, AutoGen, ou des implémentations personnalisées d'appel d'outils (tool-calling) utilisant les API natives des modèles.
  • Expertise en bases de données vectorielles : Expérience pratique des entrepôts vectoriels (Pinecone, Weaviate, pgvector, ChromaDB), incluant les stratégies d'indexation, le réglage de la recherche par similarité et les approches de récupération hybride.
  • Fine-Tuning et adaptation : Familiarité avec les méthodes de réglage fin efficaces en paramètres (LoRA, QLoRA) et savoir quand appliquer le fine-tuning vs le prompt engineering vs le RAG pour un cas d'utilisation donné.
  • Sécurité IA : Compréhension des injections de prompts (prompt injection), des vecteurs de jailbreaking et des risques de fuite de données spécifiques aux LLM — et capacité à concevoir des mesures d'atténuation au niveau de l'architecture.
  • Architecture système : Capacité à concevoir des workflows techniques évolutifs capables de gérer efficacement un volume élevé de requêtes IA.
  • Résolution de problèmes : Forte capacité à déboguer les comportements de modèles « boîte noire » et à itérer jusqu'à ce que les seuils de précision soient atteints.
  • Certifications pertinentes : Un atout.


Rejoindre Datmos, c'est :

Atteindre votre plein potentiel :

  • Vous challenger : en travaillant auprès de grands groupes et marques reconnues en  Amérique du Nord (i.e : Group CH Five Below, Skullcandy, Yeti Cycles, Casper, Jackson Mountain Resort, MasterCraft,  et bien d’autres)  ainsi qu’auprès de partenaires de renom dans le digital tels que Bloomreach, Shopify, Big Commerce, Akeneo, Klaviyo, Adobe, etc.
  • Vous surpasser en développant votre expertise : Accédez  à toutes les certifications que vous souhaitez et formez-vous tout en étant payé.
  • Contribuer à l'évolution de notre vision : apportez vos idées innovantes et votre regard d'expert. 
  • Apprendre des meilleurs : venez travailler avec plus de 100 experts certifiés dans l'une des agences les plus reconnues du marché.


Mettre votre bien-être en priorité :

  • Accès à des assurances santé payées en partie par l’employeur (assurances pour vous et votre famille).
  • Vacances et jours mobiles pour vous reposer.
  • Politique de déconnexion pour favoriser l'équilibre vie professionnelle/vie personnelle.
  • Flexibilité quant à votre lieu de travail et vos horaires (bureaux aux États-Unis et au Canada, possibilité de faire du télétravail à 100%).
  • Un programme de reconnaissance sous forme de cartes-cadeaux ou de dons à des associations.


*Datmos fait partie des meilleurs lieux de travail par Great Place to Work Canada depuis 2022.

N'attendez plus et rejoignez l'équipe !

Company Information

Company Name: O2 Commerce

Company Website: https://www.o2commerce.com

Company Address: São Paulo, SP, Brazil

O2 Commerce is a technology company that provides cloud-based commerce infrastructure and services designed to enable retailers, brands, distributors and marketplace sellers to operate and scale omnichannel digital commerce operations. The company’s core offering is a modular, API-first commerce platform that unifies order management, inventory orchestration, marketplace connectivity and fulfillment workflows so customers can sell consistently across web storefronts, mobile apps, social channels, marketplaces and physical stores. O2 Commerce positions itself as a technology partner for mid-market and enterprise merchants that need to coordinate complex selling scenarios (multi-warehouse inventory, distributed fulfillment, multi-country tax and duty calculation, and marketplace catalog syndication) without replacing existing back-office systems. At the center of O2 Commerce’s product set is an order management and orchestration layer that ingests orders from multiple sales channels, applies business rules for routing and allocation, and coordinates downstream fulfillment and shipping. This order management capability typically includes features for real-time inventory visibility (across warehouses, stores and drop-ship partners), reservation and allocation logic (backorders, partial shipments, and multi-source allocation), returns and exchanges orchestration, and the ability to generate consolidated shipping documentation and carrier labels. The platform’s orchestration engine is intended to reduce manual intervention in complex fulfillment paths and to provide a single operational view of order state across channels. Another major component of O2 Commerce’s solution is marketplace and channel integration. The company provides pre-built connectors and integration frameworks to synchronize product catalogs, pricing, inventory levels and orders with major online marketplaces and commerce endpoints. By centralizing catalog and inventory management, brands and retailers can maintain consistent product information and availability across Amazon, eBay, Walmart, social commerce channels and regional marketplaces without needing separate integrations for each channel. For sellers that use multiple marketplaces and regional storefronts, O2 Commerce’s channel adapters aim to reduce the engineering overhead of maintaining many point-to-point integrations. O2 Commerce also offers inventory and fulfillment features tailored to modern distributed supply chains. Functionality commonly described in the company’s materials includes warehouse management primitives, support for multi-node fulfillment (DCs, stores, third-party logistics providers), drop-ship orchestration, batch and work order management, and rules-based routing that optimizes cost, delivery time and inventory utilization. The platform’s approach to fulfillment focuses on enabling blended fulfillment models — for example, ship-from-store, buy-online-pickup-in-store (BOPIS), and ship-to-store — by exposing real-time store inventory and enabling rapid reservation and fulfillment decisions. The platform is delivered as a cloud-native SaaS product that emphasizes extensibility and integration. An API-first architecture and webhooks enable merchants and system integrators to build custom storefronts, rich headless commerce experiences, and to integrate O2 Commerce with ERPs, accounting systems and third-party logistics platforms. Complementary services typically offered include implementation and integration services, professional services for systems design, managed services for platform operations, and training and support engagement models to help customers move to production quickly and operate the platform at scale. In addition to core commerce and fulfillment capabilities, O2 Commerce provides analytics and operational dashboards that surface order lifecycle metrics, inventory health, fulfillment KPIs and channel performance. These analytics are designed to aid merchandising and operations teams in decision-making: identifying stockouts, optimizing replenishment, monitoring marketplace performance, and analyzing cost-to-serve across different fulfillment strategies. The company emphasizes visibility and control as primary benefits for customers seeking to reduce stock duplication, lower shipping costs and improve on-time delivery performance. O2 Commerce’s customer base, as described in its public materials, generally includes omnichannel retailers, direct-to-consumer brands, multi-brand distributors and third-party sellers that participate in marketplaces. The platform is particularly oriented toward organizations that operate across several selling channels and geographies and require a centralized operational model to ensure consistent product and order data, streamline fulfillment, and manage returns effectively. Typical use cases include enabling rapid marketplace expansion, operationalizing BOPIS and ship-from-store programs, migrating legacy OMS and integration stacks to a modern cloud architecture, and consolidating multiple storefronts and international feeds into a single control plane. Security, compliance and operational resilience are areas O2 Commerce highlights as part of delivering an enterprise-grade SaaS offering. The platform supports role-based access controls, audit logging, and secure integrations with payment processors and carrier networks. For customers operating across borders, the solution can be configured to account for taxes, duties and shipping rules specific to regions, although exact functionality varies by deployment and the integrations chosen by the customer. O2 Commerce typically positions itself as a technology-first partner that balances out-of-the-box commerce workflows with the flexibility required by enterprises to implement custom business rules. The company’s go-to-market approach integrates direct sales, partnerships with systems integrators and commerce agencies, and reseller or marketplace channel partnerships to reach retailers and brands. Professional services and partner ecosystems are used to accelerate implementations, especially where complex ERP or warehouse automation integrations are required. Overall, O2 Commerce is presented as a commerce technology provider focused on simplifying multi-channel order and fulfillment complexity through a centralized, API-driven platform. The company’s products and services target retail and distribution organizations seeking to modernize commerce operations, expand into marketplaces and optimize omnichannel fulfillment while preserving connections to existing enterprise systems.
Visit company vacancy