--- title: Data and Environment Curation for Post-Training LLMs —... description: Mahesh Sathiamoorthy, co-founder of Bespoke Labs, presents research on data and environment curation for post-training LLMs. The talk covers the Open Thoughts... canonical: https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs | daily.dev og:description: Mahesh Sathiamoorthy, co-founder of Bespoke Labs, presents research on data and environment curation for post-training LLMs. The talk covers the Open Thoughts... og:url: https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50 og:image: https://api.daily.dev/og/posts/C4ppQDf50.png og:image:alt: Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 19 min read · 0 upvotes · 0 comments ## Summary Mahesh Sathiamoorthy, co-founder of Bespoke Labs, presents research on data and environment curation for post-training LLMs. The talk covers the Open Thoughts project — a reasoning dataset created in collaboration with Stanford, UC Berkeley, and others — detailing the curation pipeline: sourcing prompts, mixing datasets, filtering questions, generating answers with teacher models, and running ablations. Key counterintuitive findings include: sampling multiple answers per question outperforms more questions answered once, and stronger models are not always better teachers. The talk extends to Open Thoughts Agents, applying similar curation principles to RL environments for agent training, where SFT still contributes significantly before RL adds the final gains. A concrete enterprise case study with Credit Karma shows how post-training improved compliance, reduced latency, and lowered costs. The Curator tooling and a broader reference stack for building RL environments and post-training agents are also introduced. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Solving the Human Training Data Problem](https://daily.dev/posts/solving-the-human-training-data-problem-n7dmfts14) · Towards Data Science · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs","url":"https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50"},"datePublished":"2026-07-31T22:01:49.721Z","dateModified":"2026-07-31T22:27:40.391Z","description":"Mahesh Sathiamoorthy, co-founder of Bespoke Labs, presents research on data and environment curation for post-training LLMs. The talk covers the Open Thoughts...","image":"https://i.ytimg.com/vi/ewtOo0scUh0/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/ewtOo0scUh0/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/data-and-environment-curation-for-post-training-llms-mahesh-sathiamoorthy-bespoke-labs-c4ppqdf50","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,reinforcement-learning","timeRequired":"PT19M","video":{"@type":"VideoObject","name":"Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs","description":"Mahesh Sathiamoorthy, co-founder of Bespoke Labs, presents research on data and environment curation for post-training LLMs. The talk covers the Open Thoughts...","thumbnailUrl":"https://i.ytimg.com/vi/ewtOo0scUh0/sddefault.jpg","uploadDate":"2026-07-31T22:01:49.721Z","duration":"PT19M","url":"https://api.daily.dev/r/C4ppQDf50","embedUrl":"https://www.youtube.com/embed/ewtOo0scUh0"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"Data and Environment Curation for Post-Training LLMs — Mahesh Sathiamoorthy, Bespoke Labs"}]} ```