Wrick Photo
Wrick Talukdar

Wrick is a distinguished AI/ML architect and product leader with over two decades of experience driving technological transformations in global enterprises. Renowned for his expertise in AI, Machine Learning, and Cloud Computing, he has consistently leveraged cutting-edge technologies to deliver strategic business value for customers all over the globe.

Throughout his career, Wrick has led digital initiatives that drive significant cost savings and accelerate time-to-market. He has a proven track record of building high-performance teams that deliver innovative AI products and solutions.

As a strategic thinker, Wrick excels in aligning technology with business goals, fostering cross-functional collaboration, and addressing complex challenges with data-driven approaches. His ability to bridge technical expertise with business vision positions him as a key leader in driving sustainable growth and innovation. Wrick is a TOGAF® Level 2 Certified Professional, holding multiple AWS and Azure certifications in Artificial Intelligence and Machine Learning.

Beyond his professional achievements, Wrick has made a lasting impact on the global scientific community, with his research widely referenced and cited worldwide. He serves as Chair of IEEE NIC, is a Senior Member of IEEE, and holds the role of Chief AI/ML Architect for Generative AI Initiatives within the IEEE Industry Engagement Committee (IEC), where he empowers young professionals to advance their careers through technology and generative AI. Additionally, as a Senior Member and Advisor of the Consumer Technology Society (CTSoc), he plays a key role in shaping technology standards and provides strategic guidance on major global conferences, including ICCE.

Projects

Agentic System for Emergency Medical Service

In critical medical emergencies, every second is vital. Harness the power of Generative AI and agentic systems with Medi-Aid to streamline decision-making and save lives. Empower healthcare professionals with real-time insights and rapid response capabilities to deliver timely and effective care.

Generate Competitive Intelligence

Leverage advanced generative AI and agentic systems to produce actionable competitive insights that drive business strategies. Unlock deeper market understanding, identify emerging trends, and make informed decisions with cutting-edge technology.

Recycle Predictor using advanced Computer Vision

Help protect our planet by leveraging advanced technology to accurately predict recyclable objects with confidence. Utilize cutting-edge AI solutions to promote sustainability and reduce environmental impact through smarter waste management.

Agentic ChatBot

Discover the power of intelligent communication with a smart ChatBot built on Agentic architecture and powered by Amazon Bedrock. This innovative solution delivers seamless, context-aware interactions, enabling enhanced user experiences and efficient task handling.

Build Safe and Secured Generative AI Applications

Leverage a robust framework designed to create safe, secure, and cost-effective generative AI applications. This solution prioritizes ethical AI practices, ensuring compliance, reliability, and scalability while empowering businesses to innovate confidently.

Live Meeting Assistant

Harness the power of advanced Automatic Speech Recognition and Generative AI models to generate real-time meeting transcripts effortlessly. This cutting-edge solution enhances productivity and ensures seamless communication, capturing every detail with precision and speed.

Clinical Notes Generator

Leverage advanced machine learning models to generate clinical notes quickly and accurately. This innovative solution streamlines documentation, reduces administrative burdens, and allows healthcare professionals to focus more on patient care.

Save clinician burnout with GenerativeAI

Effortlessly generate SOAP and BIRP notes using a robust and ready-to-use framework. This solution simplifies clinical documentation, improves workflow efficiency, and ensures consistency in maintaining detailed and accurate patient records.

Recognize faces and human emotions

Quickly and accurately recognize faces using state-of-the-art machine learning and computer vision techniques. This solution enhances security, personalization, and accessibility, providing powerful capabilities for a wide range of applications.

Create business presentations using Generative AI

Create professional business presentations in seconds by leveraging advanced Generative AI techniques and tools. This innovative solution streamlines the presentation creation process, allowing you to quickly generate tailored, high-quality slides that meet your specific needs.

Break the language barrier

Break language barriers and communicate effortlessly with anyone around the globe using TalkLocal. This powerful solution enables real-time translation, allowing you to converse in any language with ease, fostering seamless connections across cultures and regions.

Research

Synthetic Data Generation for Improving Clinical Documentation

Accurate and comprehensive clinical documentation is crucial for delivering high-quality healthcare, facilitating effective communication among providers, and ensuring compliance with regulatory requirements. Through extensive experiments on a large dataset of anonymized clinical transcripts, we demonstrate the effectiveness of our approach in generating high-quality synthetic transcripts that closely resemble real-world data.

Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite As: arXiv:2406.06569 [cs.CL] Journal: International Journal of Innovative Science and Research Technology: Vol. 9 (2024): No. 5, 1553-1566
Read paper

A Cost-Effective Approach of Classifying Financial Documents with Vector Sampling using Multi-modal Embedding Models

Accurate classification of multi-modal financial documents, containing text, tables, charts, and images, is crucial but challenging. Traditional text-based approaches often fail to capture the complex multi-modal nature of these documents. We propose FinEmbedDiff, a cost-effective vector sampling method that leverages pre-trained multi-modal embedding models to classify financial documents. Our approach generates multi-modal embedding vectors for documents, and compares new documents with pre-computed class embeddings using vector similarity measures.

Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) Cite As: arXiv:2406.01618 [cs.IR] Journal: International Research Journal of Modernization in Engineering Technology and Science: Vol. 06 (2024): No. 5, 6142-6152
Read paper

Synergizing Unsupervised and Supervised Learning: A Hybrid Approach for Accurate Natural Language Task Modeling

While supervised learning models have shown remarkable performance in various natural language processing (NLP) tasks, their success heavily relies on the availability of large-scale labeled datasets, which can be costly and time-consuming to obtain. This paper presents a novel hybrid approach that synergizes unsupervised and supervised learning to improve the accuracy of NLP task modeling.

Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG) Cite As: arXiv:2406.01096 [cs.CL] Journal: International Journal of Innovative Science and Research Technology: Vol. 9 (2024): No. 5, 1499-1508
Read paper

LLMs can be highly effective in healthcare

Comprehensive clinical documentation is crucial for effective healthcare delivery, yet it poses a significant burden on healthcare professionals, leading to burnout, increased medical errors, and compromised patient safety. We present a case study demonstrating the application of natural language processing (NLP) and automatic speech recognition (ASR) technologies to transcribe patient-clinician interactions, coupled with advanced prompting techniques to generate draft clinical notes using large language models (LLMs).

Subjects: Artificial Intelligence (cs.AI) Cite As: arXiv:2405.18346 [cs.AI] Journal: International Journal of Innovative Science and Research Technology: Vol. 9 (2024): No. 5, 994-1008
Read paper

Data extraction using multi-modal LLMs

Multi-modal large language models (LLMs) have shown remarkable performance in various natural language processing tasks, including data extraction from documents. However, the accuracy of these models can be significantly affected by document in-plane rotation, also known as skew, a common issue in real-world scenarios for scanned documents. This study investigates the impact of document skew on the data extraction accuracy of three state-of-the-art multi-modal LLMs: Anthropic Claude V3 Sonnet, GPT-4-Turbo, and Llava:v1.6. We focus on extracting specific entities from synthetically generated sample documents with varying degrees of skewness.

Subjects: Computation and Language (cs.CL); Information Retrieval (cs.IR) Cite As: arXiv:2406.10295 [cs.CL] Journal: Journal of Artificial Intelligence Research: Vol. 4 (2024): No. 1, 176-195
Read paper

Context aware grounding can improve LLM's fidelity

As Large Language Models (LLMs) become increasingly sophisticated and ubiquitous in natural language processing (NLP) applications, ensuring their robustness, trustworthiness, and alignment with human values has become a critical challenge. This paper presents a novel framework for contextual grounding in textual models, with a particular emphasis on the Context Representation stage.

Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite As: arXiv:2408.04023 [cs.CL] Journal: World Journal of Advanced Engineering Technology and Sciences, 2023, 10(2), 283-296
Read paper

Trust, safety, and ethics in development of LLMs

Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite As: https://doi.org/10.55662/JST.2023.4605 Journal: https://thesciencebrigade.com/jst/article/view/245/237
Read paper

Publications

Security in Agentic and Multi-agent Systems

As agentic systems evolve, their increasing complexity introduces significant security vulnerabilities that require immediate and proactive attention.

Read More

Reinforcement Learning in Agentic Systems

Reinforcement Learning has emerged as a cornerstone of modern artificial intelligence, enabling systems to learn optimal strategies through interaction with their environments. This can shift industries and change the way AI systems work.

Read More

Rise of Agentic AI across industires

Agentic AI and multi-agent systems are revolutionizing industries, transforming operations, enhancing efficiency, and driving innovation beyond theory into real-world impact.

Read More

Autonomous AI Agents for adaptive decision making

AI-powered autonomous agents, driven by LLMs are transforming industries by enabling systems that learn, reason, and act independently. This shift marks a move from traditional tools to intelligent partnerships, reshaping how we live, work, and interact.

Read More

Agentic Systems for Competitive Intelligence

Explore the transformative role of Agentic systems in Competitive Intelligence, generating business insights.

Read More

How Agentic Systems can save life

Agentic systems can be highly effective in emergency situations. Such systems can be life changing while dealign with healthcare emergency.

Read More

Improving Large Language Model (LLM) fidelity through context-aware grounding

A novel framework for contextual grounding in Large Language Models (LLMs)

Read More

Enhancing Clinical Documentation with Large Langauge Models

Deliver high-quality clinical documentation with machine learning

Read More

Synergizing Unsupervised and Supervised Learning

Novel hybrid approach that synergizes unsupervised and supervised learning to improve the accuracy of task modeling

Read More

Transform insurance risk assessment with agents

Enhance risk assessment and evaluate risk factors in real-time using agentic systems.

Read More

Harnessing Generative AI for Patient-Centric Clinical Note Generation

Use Generative AI and Automatic Speech Recognition(ASR) to generate highly accurate clinical notes

Read More

Cost-effective approach with multi-modal embedding

a cost-effective vector sampling method that leverages pre-trained multi-modal embedding models to classify financial documents

Read More

Elevate Healthcare documentation with Generative AI

Use generative AI (Artificial Intelligence) to produce clinical notes and enhance the quality of clinical documentation

Read More

Transform global speech into local language

Generate local language and subtitle from any languages in the world using large language models(LLMs)

Read More

Impact of document skew on data extraction accuracy with LLMs

Alternative approaches with multi-modal model architecture to build robust extraction pipeline using Large Language Models(LLMs)

Read More

Guardrails for trust, safety, and ethical development of Large Language Models(LLms)

Implement safeguards and guardrails ensure that the content generated by LLMs are safe, secure, and ethical.

Read More

Use Comprehend to ensure privacy and safety of LLMs

Enhance the overall safety and privacy of AI application using Amazon Comprehend.

Read More

Enterprise grade Natural Language Pipeline

Build a classification pipeline easily using the simplified solution.

Read More

Handle customer objections efficiently using AI

Enhance customer experience easily using efficient machine learning objection handling techniques.

Read More

Use Computer Vision(CV) to enhance extraction

Train bespoke document classification models on native documents that support layout in addition to text, increasing the accuracy of the results.

Read More

Intelligent document processing

Use advanced machine learning techniques and computer vision to process millions of documents efficiently with high accuracy.

Read More

Understand targeted sentiment with machine learning

Enable accurate and scalable brand and competitor insights using artificial intelligence.

Read More

Get critical insight from customers using machine learning

Extract meaningful information from product reviews, analyze it to understand how users of different demographics are reacting to products, and discover aggregated information on user affinity towards a product.

Read More

Extract key information from identity documents

Automatically extract information from identification documents, such as driver’s licenses and passports.

Read More

Natural Language Understanding

Build conversational experience using Lex.

Read More

📚 My Books

Building Agentic AI Systems

Building Agentic AI Systems

Create intelligent, autonomous AI agents that can reason, plan, and adapt to real-world challenges.

Pre-order now

Comprehensive Guide

A deep exploration of the key enterprise challenges in generative AI that are frequently overlooked. Expected Fall 2025.

🚧 In Progress

Professional Memberships

Senior Member

Institute of Electrical and Electronics Engineers(IEEE, USA)

Active Contributor

The Open Worldwide Application Security Project(OWASP). Top 10 for Large Language Model Applications

Advisory Member (Artificial Intelligence)

IEEE Industry Engagement Committee(IEC)

Senior Member

IEEE Consumer Technology

Senior Member

Computational Intelligence

© 2024 Wrick Talukdar. All rights reserved.