elyadata

When it comes to knowledge, everything is relevant. Especially when our role goes beyond technology… #FridaySession At elyadata, we believe that understanding our clients’ industries is just as important as mastering the technology we bring to the table. During one of our recent Knowledge Sharing sessions, our software data engineer Chahnez Chouba took us beyond the data, into the intricate world of insurance. By exploring the mechanisms behind risk management, pricing dynamics and the policy lifecycle, our teams are sharpening their understanding of a complex, high-stakes sector. This deep dive is part of our commitment to stand alongside our partners, not just as technology experts, but as collaborators who understand their business challenges and are ready to co-create impactful, tailored solutions.

Final Episode: The time for truth Building a tech company while living inside it is a unique kind of adventure... The walls are still rising, the roof isn't complete and yet life inside goes on. Teams evolve, people grow and through it all, we continue to build together. In this final episode of our Engineering Culture series, we’re sharing the unfiltered truth: ✅ What worked. ❌ What didn’t. And what we’ve learned along the way. This is a story about cohabitation, where personal ambition meets collective vision and where resilience, adaptability and shared purpose shape the future. https://lnkd.in/dpvbpnHN #EngineeringCulture #Leadership #Teamwork #GrowthMindset #PeopleFirst #TechLeadership

80% des projets IA échouent. Selon une étude de Gartner, les difficultés viennent rarement d’un manque de compétences techniques. Ce sont surtout les défis humains qui complexifient la réussite : des visions qui ne sont pas toujours alignées entre les différentes parties prenantes, des attentes parfois décalées face à ce que l’IA peut réellement apporter et une difficulté naturelle à intégrer ces solutions dans le quotidien des équipes métiers. C’est normal : l’IA est encore une technologie nouvelle, souvent entourée d’idées reçues ou d’attentes élevées. Chez elyadata, on est convaincus que la clé c’est la co-construction. 🟢 Associer les métiers et les décideurs dès le début 🟢 Travailler ensemble sur une vision commune, 🟢 Puis concevoir des solutions utiles et adoptées sur le terrain. Parce qu’une IA réussie, ce n’est pas juste de la technologie. C’est un projet collectif, ancré dans la réalité et porté par ceux qui vont l’utiliser. #IA #IntelligenceArtificielle #DéploiementIA #IAenAction

Dans cette interview, Salim Jouili, Ph.D. nous plonge dans la genèse de l'entreprise et partage sa vision sur la data, l'IA et l'humain. Un récit authentique qui dévoile ce qui fait vraiment la différence d'elyadata. #interview #IA #humanité À découvrir ici : https://lnkd.in/eFdGRX-m

Easy Setup, No Hassle #KnowledgeSharing Setting up workshop environments can be a handful. Picture hosting a hands-on session on containerization, cloud infrastructure, or DevOps, ensuring participants have the right setup without compatibility issues or lengthy configurations is often a challenge. In his Friday Session, our Software Engineer Nader AYED introduced elyalab, a platform that simplifies access to isolated, customizable virtual lab environments. The Challenge: Traditional approaches come with limitations: -Local setups: Full control and privacy but require manual configuration, come with resource constraints, and can lead to compatibility issues. -Pre-configured platforms: Quick to launch but often lack flexibility, customization options, and strict isolation. An Alternative Approach: The goal of elyalab is to provide a controlled, virtualized workspace that balances flexibility, security, and ease of access: ✅ One-click access to pre-configured environments, generate a link, share it, and participants can start working instantly. ✅ Virtualized execution, supporting container-based isolation such Docker with Sysbox or gVisor and other techniques. ✅ Automated provisioning, dynamically creating environments as needed through a Lab Engine. ✅ Secure multi-user access, preventing interference between workspaces while maintaining a controlled setup. ✅ Support for various virtualization options, making it adaptable to different use cases. With this approach, setting up an environment becomes an easy process rather than a hurdle, giving less time to setups and more time to hands-on work! #VirtualLabs #VirtualEnvironments 

AI-related jargon keeps getting more complex... 😵 Just when you think you've mastered one term, a dozen more pop up! That’s why we created The AI'phabet: an A to Z guide for those who want to understand AI, those who think they understand but might not, and those who already get it but want to explain it in simpler terms. Have fun while learning! 🤓 And if you have any terms in mind, feel free to share in the comments, and we'll give you a fun and accessible definition! #KnoweldgeSharing #AI #AttentionMechanism

Rethinking insurance, a story of transformation Insurance has always balanced precision, risk and pricing. But how do traditional models adapt to today's complexities? In this exclusive LinkedIn series, we take you behind the scenes of our collaboration with a leading insurer, revealing how our team tackled real challenges from pending data to evolving risk landscapes. Each episode shows our approach and the tangible results we delivered. Check out our first episode: "The Anticipation Strategy" and stay tuned for more! #InsurancePricing #AITransformation #DataTransformation #EngineersStories

🤔

Plot Twist: These Podcast hosts Aren't Human! Dive into our comprehensive case study as it comes to life through two AI-native hosts who expertly discuss the future of underwriting and insurance. They navigate industry transformations, technological challenges and concrete solutions in a whole new podcast format. Are you ready to discover what AI can reveal about the future of insurance? Listen now! #Insurance #underwriting #AI #businesstransformation #HumanAndTech

The Role of Rate Limiting in API Security. #KnowledgeSharing #FridaySession Ending the week on an insightful note, our Software Engineer, Azer LARIBI, took the stage to discuss rate limiting, a vital technique in API security that protects systems by regulating the number of requests a server can handle within a specific time frame. This topic was selected to highlight the significant risks of operating without rate limiting, such as: 🟥Denial of Service (DoS): Overwhelming a server with excessive requests to disrupt services. 🟥Brute Force Attacks: Repeated attempts to guess passwords or access keys. 🟥Resource Exhaustion: Overloading servers by consuming memory, CPU, or bandwidth. Rate limits act as a gatekeeper in these situations, controlling traffic and ensuring fair resource distribution. When a client exceeds the defined limit, the system responds with an error message (HTTP status code 429 - Too Many Requests). Azer explored how rate limits work and introduced the key algorithms that make them effective, such as: 🟩Fixed Window: Counts requests within a set time period and resets at the end of that period. Exceeding the limit blocks further requests until the window resets. 🟩Sliding Window: Continuously monitors requests within a rolling time frame, adapting as older requests fall outside the window. 🟩Token Bucket: Uses a bucket of tokens, where each request consumes one token. Tokens refill at a steady pace, allowing bursts of traffic when tokens are available. 🟩Leaky Bucket: Processes requests at a consistent rate, discarding any that exceed the bucket’s capacity. Although rate limiting offers a strong defense against API misuse and resource overload, it presents its own set of challenges. Striking the right balance between security and usability is essential to its success. #SoftwareEngineering #API #Security