Oct 22, 2026 · 27 min
Moshi turns market overload into trader-specific recommendations
The Recommendation Engine for Commodity Traders — Anekha Sokhal, Moshi
The episode examines whether recommendation systems can make fragmented, sensitive market information useful inside real trading workflows.
- 1Moshi combines public and private data to generate workflow-specific insights for commodity, energy, and macro traders.
- 2Trusted sources, divergent signals, and confidence scores help traders distinguish meaningful developments from noisy market information.
- 3Building AI products requires balancing third-party infrastructure, privacy constraints, customer discovery, and the practical limits of small teams.
Don't miss
Sokhal explains how Moshi uses trusted sources, divergent signals, and confidence scores to help traders judge which market developments matter.
The brief
Anekha Sokhal’s path runs from mathematics and quantitative trading to founding Moshi, after experiments in computer vision convinced her to build rather than trade around technology.
Moshi targets commodity, energy, and macro traders by combining public and private data, trusted sources, and conversational AI into recommendations tied to each user’s portfolio.
The central problem is filtering: traders face hundreds of new information pieces in seconds, while connected graphs can expose second- and third-order effects that ordinary news streams miss.
The conversation also turns practical, weighing third-party models against internal systems when customer data is sensitive and engineering capacity is limited.
Conference interest reveals the business challenge behind the product: strong inbound demand creates both customer opportunities and hiring noise, making higher-signal discovery essential.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.