Abhishek Kumar

#GenAI #MachineLearning #DeepLearning #AIeducation #BinaryCrossEntropy #MLE #AIbasics #DataScience #LLM #MathForAI #PlakshaUniversity #AIresearch #Plaksha 🚀 Introducing: Math Nuggets for Gen AI 🚀 I’m excited to kick off a new series where I break down core mathematical concepts behind Generative AI, making them simple and actionable for AI practitioners, researchers, and enthusiasts. This series is dedicated to my Father and Math Guru Prof Rajkumar Roychoudhury - I am yet to take over his Math Books and get them Republished! 🔹 Post 1: Binary Cross-Entropy (BCE) Loss Understanding BCE is crucial because it forms the foundation of many AI models, including Logistic Regression, Neural Networks, and Deep Learning classifiers. It’s not just a loss function — it’s deeply connected to Maximum Likelihood Estimation (MLE), making it mathematically elegant and powerful. 📘 In this first nugget, we explore: What BCE really represents Why it’s used for classification problems Its direct link to maximizing model likelihood Follow this series to strengthen your AI fundamentals — one concept at a time. Amrik Sen, Pankaj Pansari, Dr. Indranil Mitra, PhD, MBA, IPR, F- RSS (UK), FRSA (UK), Rajarshi Sengupta, Rajarshi Ghosh,Dr Chiranjiv Roy, PhD, MBA, Ashoka Choudhury, Abhijit Dasgupta, Sumit Ghosh, Amit Ojha, Chaitanya Repaka, Jyotirmoy Dutta, Avishek Mukhopadhyay, Rashpinder Kaur, Subhra Bishnu, Kushal Shah, Bishal Neogi, Nilanjan Saha, Shirin Khosravi Jam, Tom Yeh, Ranajoy Chaudhury, Ph.D.

118

16 Comments

With the rapid rise of GenAI and Agentic AI trainings, it can be challenging to identify programs that truly balance theory with real-world application. Recently, my Data Science team at Target In India and I attended an abridged two-day Agentic AI training led by Abhijith Neerkaje and Ajay Shenoy —both former colleagues whose deep retail and data science expertise significantly elevated the experience. Drawing from their hands-on leadership of retail data science teams, they clearly explained core concepts while grounding them in practical use cases such as supply chain and pricing optimization. The sessions were engaging, practical, and rich with insights directly applicable to live projects. Their clarity, patience, and focus on real-world relevance stood out. I highly recommend their full Agentic and Generative AI programs to anyone looking to effectively bridge theory and industry application.

45

1 Comment

Having worked with a wide variety of LLMs, this is a great roadmap to follow for anyone who wants to truly learn NLP

9

Quantization in Large Language Models (LLMs) As LLMs scale, intelligence is no longer the bottleneck — memory, latency, and cost are. Quantization turns out to be one of the most powerful (and misunderstood) techniques enabling real-world deployment of large models. In this blog, I break down: • Why LLMs are surprisingly quantization-friendly • INT8 vs INT4 vs NF4 — what actually works in practice • How modern architectures absorb low-bit noise • Where quantization fails and how to design around it Quantization is no longer an optimization step — it’s foundational infrastructure for production LLM systems. Link: https://lnkd.in/gqGje75G #LLMs #AIEngineering #Quantization #MachineLearning #GenAI #MLOps

9

2 Comments

Excited to share that my peer-reviewed work, “Enhancing BRepMFR for Machining Feature Recognition via Vision Transformers and Geometry-Aware Fourier Positional Encoding,” has been accepted for presentation at the ASA 2026 Symposium on Data Science and Statistics. This work explores how Vision Transformers combined with geometry-aware Fourier positional encoding can enhance B-Rep CAD representations using UV-grid sampling within Graph Neural Networks (GNN), enabling improved machining feature recognition directly from STEP files. A special thanks to Dr. Richard Leu for guidance and support. Looking forward to presenting and engaging with the community. If you’re attending the symposium, feel free to stop by the presentation. I’d love to connect and discuss!

25

1 Comment

IndiaAI Mission Struggles to Keep Pace Globally IndiaAI continues to face hurdles in building a competitive homegrown LLM amid intensifying international competition. While the mission focuses on compute infrastructure, safety frameworks, and indigenous efforts such as Sarvam AI’s model, experts argue that significantly more investment and capacity are required for India to lead in AI innovation. #Google #CleverTap

4

Wrote a blog explaining "LoRA Fine-tuning Efficiency Under Different Loss Functions." LoRA made fine-tuning cheap, but not always efficient. This post explores how loss functions like Cross-Entropy, Label Smoothing, and Focal Loss change convergence, stability, and generalization in lightweight LLMs. Read here: https://lnkd.in/gh_vm2kG

2

🌟 From curiosity to stage: my journey with DSPy at The Fifth Elephant A week ago, I had the incredible opportunity to speak at Hasgeek’s Fifth Elephant Conference — one of India’s most vibrant gatherings for data and AI enthusiasts. When I first started exploring DSPy (Community), I was fascinated by how it simplifies building structured, reliable AI systems — turning messy prompt engineering into a declarative, modular process. Over time, that curiosity turned into something bigger — a belief that DSPy represents the next step in how we design AI systems. In my talk, I shared how this framework can help us: 🔹 Build reasoning-centered LLM pipelines 🔹 Reduce prompt fragility through programmatic supervision 🔹 Apply these ideas to real-world problems 🎥 If you missed the session, the full recording is now available — and I would love for you to check it out: 👉 https://lnkd.in/dRBSFRF5 Whether you are building AI agents, working with LLMs, or just exploring what’s next in applied AI — I hope this talk sparks a few new ideas. #DSPy #AI #LLM #MachineLearning #fifthel #Hasgeek #AIEngineering #LLMAgents #ShiftScheduling #OpenSource Hasgeek

73

Curious to learn how Graphs are used in diverse fields ranging from Drug discovery to Finance? Announcing the 2nd Learning on Graphs Meetup (LoG) at IIT Delhi on Sat 28th March 2026. https://lnkd.in/gCXtfGRR A day full of exciting talks, posters, and panel discussions on graphs and its applications. Further, there is an opportunity for you to present your research work too. Do join us for this exciting event. We welcome folks from both academia and industry. The meetup of Learning on Graphs Conference is jointly organized by Mastercard AI Garage, TigerGraph and Yardi School of Artificial Intelligence, IIT Delhi. Nitendra Rajput Srikanta Bedathur Sayan Ranu Aakarsh Malhotra (Ph.D.) Victor Lee Rajeev Shrivastava

25

I’m reaching out to companies building in the AI space. ACM Dallas is hosting a technical workshop on Observability & Tracing in AI Agents, and we’re opening sponsorship opportunities. This isn’t a generic AI talk. The audience will be engineers, researchers, and practitioners working on real production systems. If your company works on: AI infrastructure Monitoring platforms Developer tooling LLM frameworks Cloud / MLOps platforms This is a great opportunity to connect with a highly technical community.

12