deepsense.ai

If you’ve been caught up in the whirlwind of AI buzzwords - and let’s face it, who hasn’t - AI Agents and agentic systems are currently taking center stage. But should this be a cause for excitement? Should you consider adopting them, or is this simply a rebranding of practices you already implement? While much of the attention gravitates toward flashy applications, let’s look at how AI Agents bring value to the less glamorous yet essential tasks that often go unnoticed but are critical to everyday operations. Read more in the last issue of the newsletter of our CEO, Robert Bogucki: https://lnkd.in/gxisSA9D #AI #AIagents #LLMs #agenticsystems

We’re thrilled to collaborate with a partner like AdaCore, who champions the contributions of women in STEM and recognizes the transformative impact of their work. From Ada Lovelace, often regarded as the first computer programmer, to Grace Hopper, the visionary behind COBOL and compiler innovations, women have been at the forefront of groundbreaking advancements in software. Who are some other women pioneers in software that have inspired you?  

Querying structured data with LLMs," discussed by Michał Pstrąg, is our solution, which you can explore here: https://lnkd.in/d-G3CBQ8

If you’re on the receiving end of the AI buzzword bingo - and we almost all are - AI Agents and agentic systems are now on the top. So, should you get excited about it? Should you be implementing it? Or maybe that is just a rebranding for the stuff you should be doing anyway? :) While most of the hype centers on splashy applications, let’s explore how AI Agents breathe life into the uncool use cases - the tasks nobody dreamt about but everyone relies on. #AI #AIagents #LLMs

If you’re working with foundation models and looking for ways to improve your workflow, this session on DSPy is for you. 👉 Watch it here: https://lnkd.in/dzYZeRAd Michał Pstrąg, our Machine Learning Engineer, provides a clear, concise overview of DSPy, a popular library for automated prompt engineering. This session provides a concise overview of DSPy's features, including LM prompting, pipeline construction, and prompt optimization. Discover its ecosystem and explore the potential for developing automated, self-improving LM pipelines. 👉 Ready to learn more? https://lnkd.in/dzYZeRAd #AI #MachineLearning #DSPy #PromptEngineering #NLP

The web is no longer just a place for us to navigate—it’s becoming a space where AI-powered Web Agents do the heavy lifting for us. These aren’t your basic bots. They’re sophisticated tools like WebGPT, WebVoyager, and Agent-E, designed to interact with websites intelligently, from parsing complex content to automating tasks once thought impossible. But here’s the kicker: they’re still in their early days. 💡 What’s happening now? 💡 What are the challenges? 💡 What's the future? Explore those slides to deep-dive. #WebAgents #WebAutomation #WebVoyager #AIagent #AI #technology

We’ve seen the hype - GenAI, chatbots, copilots - and now, AI agents. But how do you separate the signal from the noise? Are these just new labels for old ideas? Below, we break down the "why" and "how" of building AI agents. We’ll show you where they fit in the matrix of chatbots, copilots, and agentic systems, and share key insights from real-world applications and R&D, including: 1. How AI agents outperform standalone LLMs in tasks like natural language database interactions. 2. Why specialization beats generalization in agentic systems. 3. Key design trade-offs to balance performance with cost and complexity. 4. The critical role of evaluation pipelines to avoid AI development pitfalls. 👉 Swipe through to discover how to build smarter, scalable AI agents for your LLM applications! Interested more? Read the full article here: 🚀 https://lnkd.in/dHrwUPhJ #AIagents #GenerativeAI #LLM #ArtificialIntelligence #AI

❓ Are frontline workers and customer-facing roles the backbone of your organization? If so, you understand how high training costs and long onboarding times can hinder productivity. But what if an AI-powered assistant could help your team hit the ground running? Our latest project showcases how LLM technology and real-time data retrieval transformed frontline operations for one of our clients. Explore our carousel to dive into the details. 👇 Curious about how LLMs, Retrieval-Augmented Generation (RAG), and SLMs on edge can revolutionize your business? Check out the comments to learn more! #GenAI #RAG #LLMs #SLMs #FrontlineWorkers #AI #Retail #Innovation

Keeping our fingers crossed for Michał Pstrąg and his talk about our framework for building robust structured data retrieval for RAG pipeline or standalone LLM app. It's called db-ally, and you can find it here: 👉 https://lnkd.in/dVhaZwUS #LLM #RAG #AI

🚨 Is your AI making things up? Stop the guesswork. Whether you're a data scientist fine-tuning models or a business leader optimizing AI workflows, this is the actionable guide you need. Why? Ever heard of an AI confidently claiming someone walked across the English Channel? Or a mental health chatbot giving dangerously irresponsible advice? These are not isolated incidents—they’re examples of LLM hallucinations, where models generate convincing but false information. If you’re deploying AI for business or technical tasks, this problem is more than an inconvenience—it’s a liability. Think about: 1. 22% of medication-related LLM answers in one study were outright dangerous; 2. high-stakes industries like healthcare or law can’t afford “I guess” moments. In our latest article written by our Senior Data Scientist Katarzyna Rutkowska, we break down how to measure and reduce hallucinations effectively: - Techniques to tackle hallucinations, from simple prompt tweaks to advanced RAG (Retrieval Augmented Generation). - Real-world tools like LLM-as-a-judge and token logprobs. - A playbook of approaches tailored to your needs—whether cost, accuracy, or speed is the priority. 👉 Dive into the full breakdown here: https://lnkd.in/dxD-7dMB #GenerativeAI #AIforBusiness #LLMHallucinations #AIDevelopment #TechLeaders #AIFailures #DataScience #LLMOptimization