Talks & workshops
Conference sessions, webinars, workshops, and community talks.
I like teaching through systems I have actually built. My sessions tend to combine a working implementation, the evaluation loop around it, and the mistakes that shaped the final approach. My longer-form courses are documented on the Work page.
Selected talks and workshops
Agent evaluations with CrewAI and W&B Weave
With Tony Kipkemboi and João Moura. A practical blueprint for tracing, evaluating, and iterating on agentic applications.
With Tony Kipkemboi and João Moura. A practical blueprint for tracing, evaluating, and iterating on agentic applications.
The Missing Piece of AI Apps: Evaluation
How evaluation changes the way teams build LLM applications—from one-off demos to systems that can improve reliably.
How evaluation changes the way teams build LLM applications—from one-off demos to systems that can improve reliably.
GenAI Workshop: RAG Zero to Hero
Two sessions on the modern GenAI landscape and building RAG systems, delivered as a hands-on community workshop.
Two sessions on the modern GenAI landscape and building RAG systems, delivered as a hands-on community workshop.
Enhancing LLMs with Structured Outputs and Function Calling
Practical patterns for making language-model outputs dependable enough to participate in larger software systems.
Practical patterns for making language-model outputs dependable enough to participate in larger software systems.
Fine-tune Command R/R+ using Cohere and W&B
With Komal Teru. Establishing baselines, tracking fine-tuning, and evaluating results for Cohere’s Command models.
With Komal Teru. Establishing baselines, tracking fine-tuning, and evaluating results for Cohere’s Command models.
Optimizing LLM apps at scale
With Bharat Ramanathan. Lessons from evaluating Wandbot and building multilingual leaderboards for real RAG systems.
With Bharat Ramanathan. Lessons from evaluating Wandbot and building multilingual leaderboards for real RAG systems.
Building flexible, evaluation-driven RAG
With Maxime Voisin. Designing RAG systems as measurable pipelines rather than fixed demos.
With Maxime Voisin. Designing RAG systems as measurable pipelines rather than fixed demos.
LLMs for the mass market
With Bharat Ramanathan and Ian Webster. Evaluation and internal-productivity lessons from deploying LLM-powered tools.
With Bharat Ramanathan and Ian Webster. Evaluation and internal-productivity lessons from deploying LLM-powered tools.
Deep dive into end-to-end MLOps using W&B
A complete workflow for tracking experiments and moving from model iteration toward reproducible ML systems.
A complete workflow for tracking experiments and moving from model iteration toward reproducible ML systems.
Deep Learning for Detection and Segmentation
An early community talk on practical computer-vision pipelines for object detection and semantic segmentation.
An early community talk on practical computer-vision pipelines for object detection and semantic segmentation.
Community
I was a Google Developer Expert in Machine Learning from 2022 to 2025. I have also taught through GDG communities, Kaggle, W&B events, and local ML groups. This page is a curated record rather than a complete calendar; I will keep adding older sessions as recordings and event pages resurface.
For a session, workshop, or collaboration, the best way to reach me is LinkedIn.