About
Activity
6K followers
-
Jaya Prasad J reposted thisJaya Prasad J reposted thisSuccess using #AI is not accidental, it's intentional. At Wolters Kluwer ELM Solutions, we've been using #AI for over 10 years to review invoices. We've poured that expertise into our #Invoice #Review #Agent which ingests outside counsel guidelines and #automatically applies adjustments, driving a more #efficient process with up to a 10x productivity gain. Our LegalVIEW BillAnalyzer Invoice Review Agent is now available - check out the latest coverage at law.com. https://lnkd.in/dqsfimiuWolters Kluwer Launches Invoice Review AI Agent for In-House Teams | Law.comWolters Kluwer Launches Invoice Review AI Agent for In-House Teams | Law.com
-
Jaya Prasad J shared thisBig step forward for the Legal industry! We’ve launched a cutting-edge Agentic AI solution powered by a workforce of 10+ specialized agents, intelligently orchestrated to transform one of the industry’s most complex and time-consuming legal processes. These agents collaborate with each other, automate work end to end, and bring a whole new level of speed, accuracy, and intelligence to legal operation.Jaya Prasad J shared thisFor years, the legal industry's approach to invoice review has been stuck in the same loop: AI flags a problem → a human reviews the flag → a human takes action. We just replaced that loop. Today, Wolters Kluwer ELM Solutions launched the LegalVIEW® BillAnalyzer Invoice Review Agent — the industry's first agentic AI that doesn't just identify non-compliant invoice line items, it autonomously acts on them: adjusting, explaining, and documenting every decision with full transparency and auditability. 📰 Full coverage from Legaltech News: https://lnkd.in/gR3cTbms This wasn't a 6-month skunkworks project. This was years of deliberate investment: 📊 $230B+ in real legal invoice data training the models 🧠 10 purpose-built AI agents orchestrated to handle everything from compliance checks to duplicate detection to adjustment calculations 👥 Hundreds of compliance and bill review experts whose decades of domain expertise are embedded in every prompt and decision layer 🏗️ Rigorous pilots with major enterprises that validated the shift from AI-assisted review to truly autonomous analysis Here's what I'm most proud of as the CPO who helped bring this to market: We didn't build an AI tool. We built AI that acts like your best reviewer — one who never misses a guideline, never gets fatigued at line 4,000, and always shows their work. And we kept humans in control. Legal teams set the guardrails. They update guidelines in natural language. Every action is reversible. Every adjustment is explainable. This is what I believe the future of legal operations looks like: Automatic where it should be. Flexible where it needs to be. Always aligned to your strategy. Enormous congratulations to our Data Science, Product, Engineering, LBA Operations, and Go-To-Market teams — and to our pilot partners who trusted us to get this right. The era of flag fatigue is over. The era of agentic action has begun. What's the one manual review process in your legal department you'd want AI to handle end-to-end? 👇 #AgenticAI #LegalOps #LegalTech #InvoiceReview #WoltersKluwer #AI #LegalInnovationWolters Kluwer Launches Invoice Review AI Agent for In-House Teams | Law.comWolters Kluwer Launches Invoice Review AI Agent for In-House Teams | Law.com
-
Jaya Prasad J shared thisWe’re hiring an AI Product Manager Come join our Product Intelligence Team to build agentic AI solutions for real-world use, transforming Enterprise Legal Management (ELM) workflows and solving real customer problems end-to-end. If you're interested, please apply using the below link https://lnkd.in/eeR5Ar2b
-
Jaya Prasad J shared thisAI at the forefront of transforming legal operations teams. Thrilled to see how far we’re pushing the boundaries!!Jaya Prasad J shared thisWe’re excited to unveil a major innovation in legal tech: single-matter summary AI functionality in TyMetrix 360°! This feature transforms how corporate legal teams manage matters and spend by instantly digesting complex data from legal invoices and matter notes. It offers: ✅ Holistic Summaries: AI consolidates line items and notes for faster collaboration and decision-making. ✅ Real-Time Snapshots: Get instant updates by timekeeper, quarter, and matter phase. ✅ Quick Reference Highlights: Access key insights at a glance. To learn more about how TyMetrix 360° is transforming legal operations, read below. #legalops #legaltech #legaloperations #lawtech #legalAIWolters Kluwer introduces new AI-powered matter summary capabilities to TyMetrix 360°Wolters Kluwer introduces new AI-powered matter summary capabilities to TyMetrix 360°
-
Jaya Prasad J liked thisJaya Prasad J liked thisWrapped up an energizing 2-day offsite with the DNA extended leadership team. We aligned on operating principles, sharpened our roadmaps, and pushed our thinking on innovation, all while having a lot of fun along the way. Most importantly, we left with clarity and conviction: FIP (FCC Intelligence Platform) is how we turn data into a true competitive asset. Onward. 🚀 Wolters Kluwer, #FCC, #FIP, #DnA Wolters Kluwer Financial & Corporate Compliance
-
Jaya Prasad J liked thisJaya Prasad J liked thisLast week, I had the opportunity to facilitate a workshop with commercial leaders at Wolters Kluwer on driving commercial excellence through Rev Ops, and share what drives our success at Brightflag. The takeaways I most wanted to leave them with: 🤝 Act in the best interest of your customers, and the results will follow 🎯 Complexity is a tax on speed; aim for simplicity in all you do ✨ In the AI era, we have a rare opportunity to stop optimizing the past and leapfrog straight to where we actually want to be Thank you to my fellow track leads Markus Lutz, MBA and Gurjinder Singh Sembhi for the collaboration, and to Martin O'Malley, Maria Joao Montenegro, and Kenneth Jakobsen for the opportunity.
-
Jaya Prasad J liked thisJaya Prasad J liked thisI walked into a 30-minute meeting with questions. I walked out lighter. Not because every question was answered. Because I felt heard. Yesterday, I met a friend who runs a highly successful consulting practice, and one thing stood out immediately: there wasn’t a single computer in his office. His team has the best tools and technology. He wants them to move faster, think better, and spend more time learning. But for himself, he made a different choice. He removed the screen. He realized that if his highest-value work was building trust, understanding nuance, and helping clients think clearly, then a monitor was competing with the very thing that mattered most: presence. That 30-minute conversation reminded me of something simple and easy to forget: People do not always leave a meeting lighter because they got solutions. Sometimes they leave lighter because they felt seen. As AI becomes part of every profession, maybe the real leadership question is this: What should technology handle — so that we can be more human where it counts?
-
Jaya Prasad J liked thisJaya Prasad J liked thisDay 01 of posting daily technical papers to advance AI skillset👇 🚀 ::Rethinking LLM Inference with Paged Attention:: As Large Language Models continue to scale, the real bottleneck in production is no longer just compute — it’s memory efficiency. One of the interesting innovations in this space is Paged Attention, introduced in the paper “Efficient Memory Management for Large Language Model Serving with PagedAttention” 🔍 ::The Problem:: When LLMs generate text, they rely heavily on KV cache to store attention information for all previously generated tokens. This reduces computational costs but comes with its own challenges: · KV cache grows dynamically with sequence length · Memory is typically allocated as large contiguous blocks · This leads to fragmentation and wastes GPU memory 👉 Result: Lower throughput, smaller batch sizes, and inefficient utilisation of expensive GPUs. 💡 ::The Solution:: Paged Attention borrows a concept straight from operating systems — virtual memory paging. Instead of allocating one large contiguous block: · KV cache is divided into fixed-size blocks (pages) · These pages are non-contiguous in physical memory · A page table maps logical sequence positions to physical blocks ⚡ ::Why It Matters:: This simple shift unlocks major gains: · ✅ Near-zero memory waste in KV cache · ✅ Efficient handling of variable-length sequences · ✅ Flexible sharing of cache across requests · ✅ 2–4× throughput improvement compared to prior systems (as reported in the paper) 📄 ::Reference:: 📘 Paper: Efficient Memory Management for Large Language Model Serving with PagedAttention 🔗 https://lnkd.in/dgPsfzBk
-
Jaya Prasad J liked thisJaya Prasad J liked this🎉 600 episodes. One mission. For 600 episodes, GAI Insights' Daily AI News has helped leaders cut through the noise and stay ahead of what matters in AI. What started as a daily briefing has become a global community of executives, practitioners, and experts learning together in real time as AI reshapes business. Tomorrow, we're celebrating this milestone alongside many of the incredible Guest Analysts who have been part of this journey. Thank you to everyone who has watched, contributed, shared insights, and made these conversations possible. Join us live as we celebrate 600 episodes and look ahead to what's next for AI. 📆 Tomorrow 🕢 Join us live tomorrow at 7:30 AM EST: https://lnkd.in/ef_ftVp2 #GAIInsights #DailyAINews #EnterpriseAI #GuestAnalysts Paul Baier John Sviokla Adam Rappaport Luda Kopeikina Ankaj Mohindroo Tim Andrews Brenda Kirk Ashish Bhatia Manik Goyal Karin Pespisa, MBA Alden Do Rosario Vivek Mukhatyar Sandeep Sacheti Kevin Petrie Nina Araujo Lucinda Linde Christopher Brock
-
Jaya Prasad J liked thisJaya Prasad J liked thisIs AI actually killing SaaS? Not the smart ones. In a recent HBR digital article, "AI's Impact on SaaS Will Be Uneven," Harvard Business School Professor Christopher Stanton debunks the "SaaSpocalypse" myth. He introduces a matrix separating software by its task (Deterministic vs. Predictive) and its data source (Internal vs. Pooled). https://lnkd.in/ek5xjYvv The true winners are in the 4th Quadrant: Operational Intelligence (Predictive + Pooled Data). These platforms solve complex edge cases by aggregating global, long-tail data signals that no single company can replicate internally. I was proud to see my employer, Cloudflare, called out as a prime example. Our ML models analyze global internet traffic (>20% of total) and threat signals. No standalone security team or single-company log can match that pooled intelligence. As Professor Stanton notes: "AI does not make all SaaS less valuable. It makes weakly differentiated SaaS easier to replace." Help me build a list: Aside from Cloudflare (cybersecurity) and Bluon (HVAC support) that professor illustrated, which B2B companies truly dominate this Predictive + Pooled Data quadrant? Where is the real value in the irreplaceable global context, not the software UI? Drop your examples below!
-
Jaya Prasad J liked thisJaya Prasad J liked thisTo Every Fresh Grad Walking Into an Interview Ran our DS internship interviews last Friday for Data Science Interns. A few honest takeaways. - The talent bar is genuinely high: Today’s graduates are easily 5x better prepared than those from five years ago. Almost every conversation was solid. The competition is real. - AI built projects are fine. AI built thinking is not: First time I heard candidates openly said no to coding questions and openly admit they used AI to build their projects. Fair enough. But when we asked why certain decisions were made, the thinking wasn’t there. That gap shows instantly. - Your CV is a contract, not a catalogue: If it’s on your CV, expect to discuss it. Overstuffed profiles hurt strong candidates more than they help. Put only what you can genuinely own. - Attitude matters as much as ability: Some candidates had strong hands-on experience, it showed. But the moment they were challenged or given feedback, the response was defensive. We hire for teams. Coachability isn’t optional. - A note to interviewers: We’ve all sat on the other side of that table. Our job is to create the conditions for someone to show their best not to filter with friction. Regardless of the outcome, every candidate should leave feeling they were genuinely heard. That’s the standard. The bar is rising fast. But so is the importance of how you think, how you carry yourself, and how you make others feel. #DataScience #Hiring #CareerAdvice #FreshGraduates #InterviewTips #AISkills
-
Jaya Prasad J liked thisJaya Prasad J liked thisIn high-stakes workflows like healthcare, tax and the law, professionals can't afford to get AI wrong. Today, we announced an expanded enterprise AI collaboration with OpenAI to bring OpenAI’s advanced AI capabilities into our existing Expert AI solutions. By combining OpenAI’s latest APIs and platform capabilities with our curated content and deep domain expertise, the clinicians, lawyers, accountants and experts we serve will be able to access more clinically relevant support and move faster through time-intensive work across legal, compliance, and regulatory workflows. We’re proud to be making this meaningful commitment to responsible, next generation AI in regulated, high-stakes fields. Read more on our website: https://ow.ly/W8qQ50Z6YiN #WoltersKluwer #OpenAI #ExpertAI #innovation
Experience & Education
-
Wolters Kluwer
View Jaya Prasad’s full experience
See their title, tenure and more.
or
Licenses & Certifications
Volunteer Experience
-
-
Treasurer
Bharathiar Youth Club
Social Services
-
Courses
-
Big Data Seminar
BIA 678
-
Data Warehousing and Business Intelligence
MIS 636
-
Experimental Design
BIA 652
-
Financial Decision Making
FIN 615
-
Knowledge Discovery and Data Mining
CS 513
-
Multivariate Data Analysis
BIA 652
-
Optimization and Process Analytics
BIA 650
-
Web Analytics
BIA 660
Projects
-
Field Oriented Speed Control of PMSM using TMS320F2812
• This project is about the speed control of a 3-phase Permanent Magnet Synchronous Motor in a turret of T-90 military tank using the Texas Instruments (TI) TMS320F2812, a digital signal processor (DSP).
• Programmed in C++ and simulated in Matlab.Other creators -
-
Smart Grids, Sensors, and Neural Networks - Optimization of Load Forecasting
-
• This project is to deal with the root cause of energy loss and to obtain a precise prediction of electricity.
• Data has been collected for New York City commercial buildings on every hour basis and converted into minute basis using Excel.
• Based on the report from Energy.gov about occupancy sensors, we developed a model which reduced the energy consumption in office buildings by 28.4%.
• Once this model is built, using the electricity consumption data on the minute basis for four…• This project is to deal with the root cause of energy loss and to obtain a precise prediction of electricity.
• Data has been collected for New York City commercial buildings on every hour basis and converted into minute basis using Excel.
• Based on the report from Energy.gov about occupancy sensors, we developed a model which reduced the energy consumption in office buildings by 28.4%.
• Once this model is built, using the electricity consumption data on the minute basis for four weeks, we built a Neural Network optimization model using Excel and obtained a prediction which gave 18.9% increase in the forecasting.Other creators -
-
Foot Charger
-
• A Mobile or low power equipment charger without using the actual power supply.
• This can be achieved by embedding Piezoelectric crystals in the sole of the shoes.
• When there is no supply of electricity, pressure can be given over the shoe, which will give out an amount of power used to charge the equipment.
• This can be used during trekking and by military people when they run out of power in their equipment. -
Hybrid electrical vehicle via Renewable Energy in train
-
• Adding to the source of energy used in "Renewable energy from moving train", one more source was implemented in this.
• Biodegradable wastes are one more source of electricity.
• This increased the efficiency of the system by 14%.
• Simulated using Matlab.Other creators -
-
Renewable Energy from Moving Train
-
• We developed this model when there was scarcity of electricity in the entire state.
• This project makes use of the renewable source of energy such as Wind and Solar to power the lights and fans in the train.
• The most interesting thing in this is the power from the wind energy, we implemented the concept of aerodynamics into the water bottles to get power from wind.
• The other mode of energy is by using solar panels on the top of the trains. To achieve more efficiency copper can…• We developed this model when there was scarcity of electricity in the entire state.
• This project makes use of the renewable source of energy such as Wind and Solar to power the lights and fans in the train.
• The most interesting thing in this is the power from the wind energy, we implemented the concept of aerodynamics into the water bottles to get power from wind.
• The other mode of energy is by using solar panels on the top of the trains. To achieve more efficiency copper can be replaced using silver.
• We simulated this using Matlab.Other creators -
Honors & Awards
-
First Prize- JNN Institute
POWER ETERNA '13 ,J.N.N Institute of Technology
• Won first prize in National level technical symposium.
-
First Prize- Anna University
ELECSTORMZ'13, Anna University
• Won first prize for paper presentation in a National level technical symposium.
-
National Conference on Green Energy
RMD Engineering college
• Presented our paper in a National conference, which had delegates from various Green energy organization.
Languages
-
English
Full professional proficiency
-
Tamil
Native or bilingual proficiency
Other similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content