deepsense.ai

This announcement from Vercel is actually worth a 2nd post because 1) it's such a big deal and 2) now I know why. The reason we've been able to build BB AI so fast is that there is ZERO communication required between business vision, product management and engineering. It's all in one person's head...mine. But all along, one of my biggest concerns was how to scale this without slowing down. We've onboarding a new engineering team to support me and their biggest roadblock is the ability to get ideas out of my head and into well documented requirements. When product can't do engineering and engineering doesn't know what the business wants, you need an effective communication layer and that just slows things down. One approach I took was to build a prototype of Jira <--> Claude (Thank you deepsense.ai for the idea: https://lnkd.in/gGhaGMYy) so that I could assign tickets to Claude in JIRA and that would spawn a cloud agent to implement the ticket, generate a PR and notify an engineer to review it. That's a start, but we need to merge the PR before a PM can review it. And if there are issues or bugs, we have to create another ticket and go back through the loop. This change from v0 / Vercel allows product managers to BUILD (not prototype) features and have engineers review / merge the code only after the feature is done properly. This massively reduces cycle time for product development. This is an ABSOLUTE GAMECHANGER. Thank you Vercel! https://lnkd.in/eeDtaGij

Imagine you’re a user of an e-commerce platform. You don’t scroll endless categories. You don’t tweak filters for 10 minutes. You just say what you’re looking for - and the platform actually understands. That’s what we helped build for a marketplace in the Gulf Cooperation Council. 👉 In 6 weeks, we moved them from static catalogs to AI-driven, conversational product discovery - designed for production. What was behind it: - A full marketplace architecture (UI → backend → partners) - LLMs embedded into personalization and discovery - Real integrations with merchants, logistics, and payments Early results: → 25% of autonomous agent actions executed correctly (above early benchmarks) → A clear path to improving reliability and scale The takeaway? Conversational commerce isn’t just chatbots. It’s about better discovery, higher conversion, and systems that can scale. This matters if you’re building: - Marketplaces or e-commerce platforms - AI features that must survive production - LLMs that work with real operations, not around them 👊 How do you imagine your ideal e-commerce experience?

Quick share for anyone working on LLMs in production 👇 We surveyed 20 CTOs and AI leaders about what’s actually happening after the PoC phase. A lot of the answers were familiar: - Great demos, painful production - ROI measured in cost + time saved, not hype - Security and compliance slowing everything down The short video sums it up well. The report goes deeper if you’re in the middle of shipping (or trying to). Worth a look if this sounds close to home: 👉 https://lnkd.in/d6h45xy9

At deepsense.ai, applied AI for business is only part of the story. Our team also actively contributes to peer-reviewed scientific research at the frontier of AI and the natural sciences. This paper presents the design and experimental validation of an LLM-based recommender system for photocatalysis - a domain where expert intuition, fragmented literature, and trial-and-error have traditionally driven decision-making. 🚀 The authors trained a transformer-based model on 36,000+ published photocatalytic reactions, enabling it to recommend suitable photocatalysts with ~90% accuracy under cross-validation. Crucially, the system was validated experimentally, where its recommendations achieved yields competitive with, and in some cases superior to, those selected by experienced human researchers. Beyond accuracy metrics, the work demonstrates something more important: LLM architectures, when grounded in high-quality domain data and paired with experimental validation, can function as practical decision-support tools for scientific discovery, not just text generators. The model is openly available and free to use, supporting reproducibility and real adoption by the research community. Authored by Michał Kulczykowski, Ph.D. and collaborators, this work reflects how our engineers and scientists operate across boundaries, applying state-of-the-art AI in production settings while advancing the scientific foundations behind it. Read the full paper: 👉 https://lnkd.in/daD-jBEw

Handling 30,000 🛑 customer emails per month, many with scanned or handwritten attachments, is not a scaling strategy - especially under strict GDPR and regulatory constraints. Together with InterRisk TU SA Vienna Insurance Group, we proved that it does not have to be this way. In a 3-week Proof of Concept, we built and validated an AI-powered platform that: - Automatically classifies incoming emails into 12 service categories - Extracts structured data from text, scans, and handwritten documents - Operates on secure, Europe-hosted infrastructure with full auditability The impact: ⏱ Processing time reduced from 5–10 minutes to ~30 seconds 📉 >90% reduction in manual effort 🎯 ~85% classification accuracy 📈 Scalable throughput of up to 5,000 emails/day 🔐 GDPR- and regulator-aligned by design (no third-party data processing) This PoC unlocked stakeholder buy-in and laid the groundwork for full production deployment and enterprise-wide automation. Read the full case study: 👉 https://lnkd.in/dszhNMZs

What if HR workflows didn’t require navigating dozens of screens in ERPs like Workday, but could be executed in plain language? For a leading global manufacturer in the mobility and tire industry, we built and validated an LLM-powered agent that automates complex, multi-step HR processes using natural language, inside a secure enterprise environment. What changed: 🔵 5 real Workday scenarios automated end-to-end 🔵 3× faster process design vs. traditional RPA 🔵 New workflows added in hours, not days 🔵 Accuracy reaching 80–100% across scenarios This project proved that LLM agents are a viable, scalable alternative to RPA for enterprise systems, reducing manual effort, lowering training costs, and making automation accessible to non-technical users. Most importantly, it laid the foundation for deterministic, production-ready automation that enterprises can trust and extend. Read the full case study to see how agentic AI is reshaping HR operations in practice: 👉 https://lnkd.in/dS8DSGeG