Large Language Model patterns and more | Abhishek Kumar (Publicis Sapient, UC Berkeley)

Abhishek Kumar

Abhishek Kumar is currently leading the AI and Data Science practice at Publicis Sapient as Director of Data Science. He leads engagements for numerous Fortune 500 companies on Data & AI strategy and large-scale implementation for areas in recommendation engines, anomaly detection, optimization, and recently in Generative AI. He holds a Master’s degree from the University of California, Berkeley, and is a recipient of the prestigious Hal Varian Award. He is also recognized as a Google Developer Expert-ML and featured in the Top 40 under 40 data scientist list. He has authored several popular data science courses and has delivered talks at global data and AI conferences. He is also a learning facilitator and instructor at Berkeley Haas Executive Education, focusing on the Future of Technology.

In this episode, we discuss:

• What is LLMOps?

• New components to the traditional MLOps infrastructure

• LLM deployment patterns

• Prompt engineering and management

• Challenges scaling LLM-powered apps

• How enterprise architecture will evolve

• Human Reinforcement Learning Feedback

• LLMOps is a team sport

• Skills required for LLMOps

• Use case and technology selection

• Trends and opportunities for LLM applications

• Change management and creative mindset

• more!

Listen now on Apple, Spotify, and YouTube.

Where to find Abhishek Kumar:

LinkedIn: https://www.linkedin.com/in/meabhishekkumar/

Where to find James Gray:

• LinkedIn: https://www.linkedin.com/in/jamesgray

• Instagram: https://www.instagram.com/coachjamesgray

• Coaching: https://coachjamesgray.com

• Data & AI Consulting: https://GrayData.ai

• The James Gray Podcast: https://graymatter.jamesgray.io/podcast

---

Referenced in the podcast

Publicis Sapient: https://www.publicissapient.com/

DataHack Summit: https://www.analyticsvidhya.com/datahack-summit-2023/

Berkeley Haas ExecEd Future of Technology: https://em-executive.berkeley.edu/future-technology

LangChain: https://www.langchain.com/

Hugging Face: https://huggingface.co/

Meta Llama2: https://ai.meta.com/llama/

Guardrails.io: https://www.guardrails.io/

Weights and Biases: https://wandb.ai/site

Anthropic: https://www.anthropic.com/

---