"AI-Powered Search" Course Testimonials

Can't wait to start signal boosting before diving deeper into LTR.

A course that covered search topics extensively, from traditional to modern methods. The amount of work that went into creating it and the ground it covers in about a month is phenomenal.

Broad overview of current techniques from leading experts. The course is especially valuable if you already have a good foundation in the area and bring your own questions to the speakers and cohort.

The course is very relevant to current industry problems in Search. I have dealt with the issues discussed in this course but lacked a direction. With this course I feel informed and confident of making suggestions to improve Search experience at my current company

AI-Powered Search by @treygrainger & @softwaredoug & AI-Powered Search: the most in-depth course I've taken for leveling up retrieval skills.

Moved me from "throw embeddings at it and hope" to actually understanding query intent, ranking signals, and production RAG patterns.

Hands-on labs + expert guest speakers from Qdrant, OpenSearch, Superlinked made it incredibly practical.

Essential for anyone serious about search & RAG. Next cohort in March 👉 https://lnkd.in/gxnkckZa

Trey and Doug have put together and delivered an excellent course that includes both breadth and depth on the topic of AI Powered Search. I appreciated how they brought some dense topics to life through examples and interactive labs. There were plenty of opportunities to ask questions throughout the lectures and they offered weekly office hours to go in depth on topics of interest to the group. I recommend this course to any search practitioners that are looking to evolve their search products with leading edge techniques. I especially recommend this course to Search Product Managers who are new to the AI-Powered Search landscape that need a grounding in the technology.

After six months of intensive AI upskilling and building RAG systems, I thought I had a decent handle on retrieval . Then I took the course AI-Powered Search from Maven , and realized how much I'd been missing. Trey Grainger and Doug Turnbull go far beyond the "download embeddings from Hugging Face and hope for the best" approach. They systematically dismantle the naive assumption that hybrid search (BM25 + cosine similarity) is "good enough." As Trey put it in the semantic query understanding session: most people just throw embeddings and keywords at the search engine and hope it comes up with the right thing. This course is about challenging that paradigm.

Just wrapped up the Maven Course: AI-Powered Search - Modern Retrieval for Humans & Agents, and wow—it's a game-changer that left me buzzing with excitement and clarity!​I dove in initially for the product/search angle, but this course delivered extreme value for technical deep-dives and business insights, empowering owners to max out search in development. The content is razor-sharp up-to-date, unpacking complex solutions like personalized behavioral embeddings, agentic search paradigms, and ML ranking with crystal-clear intuition that ties every piece together—no more black-box confusion.​What blew me away? The perfect learning intensity, permanent access to recordings/notebooks/slides, and an amazing community of top-tier lecturers/peers —plus live 'office hours' for diving into any question, even off-scope. It's not just theory; it's hands-on mastery bridging demos to production systems that actually crush relevance for humans and agents.​If you're a dev, founder, or search-curious pro, this is your edge in the AI retrieval revolution—join!

I took this course because, even after doing search work for a couple of years (mostly self-taught), I kept running into the same problem: I could research concepts online, but the pieces didn’t always click into a production-grade mental model. This class was exactly what I needed. Doug and Trey bridge the gap between theory and real systems in a way that’s hard to get from blog posts and docs, especially the practical “how” behind query understanding, making hybrid retrieval behave predictably, and turning behavioral signals into an iteration loop that actually improves ranking. The biggest value for me was the experience transfer: the nuance, the “why,” and the decision-making that comes from having built and tuned search at scale. The live format and Q&A made a huge difference, you can bring your real problems and get grounded guidance. The hands-on labs were also excellent: a reusable set of notebooks and techniques I can apply to my own stack. If you’re building search/RAG systems and feel like you’re missing the foundations (or the production intuition), this course fills that gap fast.

Excited to share that I’ve completed Maven’s AI-Powered Search course .
Over 5 weeks, I went deep on what it takes to build modern, relevance-driven search systems—from classic IR foundations to LLM-era retrieval and ranking.
A few key takeaways I’m walking away with:
Relevance is a product problem as much as a modeling problem: user behavior insights, implicit signals, and evaluation loops matter as much as architecture.
Hybrid retrieval is the default: mixing sparse + dense (e.g., MiniCOIL-style approaches) and tuning the blend for real traffic.
Ranking is the real lever: LTR, click models, bias-aware learning, and active learning to keep improving without endless labeling.
RAG done right: interleaving strategies, retrieval quality, and grounding—because “more context” ≠ “better answers.”
Performance + UX: vector search optimization, semantic query understanding, semistructured embeddings, and agentic search patterns.
Big thanks to the speakers who shared practical lessons and real-world war stories Doug Turnbull and Trey Grainger🙏
Next up: applying these ideas to AI-powered conversational search—improving relevance, grounding, and measurable user outcomes.

I took this course because we hit a ceiling. We built a RAG-based search, and for a while, we were making progress. But at some point, we were just rearranging deck chairs, getting marginal improvements while the big problems weren't moving. Latency issues, gnarly quality gaps, the things that actually mattered. I realized I needed a deeper understanding of search and retrieval to make real gains. I had previously read Doug and Trey's AI-Powered Search book and decided to take their class. I'm also a fan of Doug's blog. What I found was a much deeper exploration of the topics in the book. They cover a lot of ground in the book, but the nuances come out in the class discussions and Q&A. Doug and Trey have worked in search and information retrieval for over a decade. When anyone in the class asked a question, they could pull from a variety of experiences to give a genuinely useful answer, not a theoretical one. You find yourself leveling up just from the conversation. Before taking the class, I felt like I had exhausted my options. Now I see there's a lot more room for improvement. They gave me new tools, but they also gave me perspective. Since they've worked on high-scale, top-tier search systems, they also set expectations for what excellence actually looks like. So you know what you're aiming for, not just what to try next. If you want to get good in this area, I can't think of a better class. I don't know of anyone else with this level of expertise who's actually teaching it.

Can't wait to start signal boosting before diving deeper into LTR.

A course that covered search topics extensively, from traditional to modern methods. The amount of work that went into creating it and the ground it covers in about a month is phenomenal.

Broad overview of current techniques from leading experts. The course is especially valuable if you already have a good foundation in the area and bring your own questions to the speakers and cohort.

The course is very relevant to current industry problems in Search. I have dealt with the issues discussed in this course but lacked a direction. With this course I feel informed and confident of making suggestions to improve Search experience at my current company

AI-Powered Search by @treygrainger & @softwaredoug & AI-Powered Search: the most in-depth course I've taken for leveling up retrieval skills.

Moved me from "throw embeddings at it and hope" to actually understanding query intent, ranking signals, and production RAG patterns.

Hands-on labs + expert guest speakers from Qdrant, OpenSearch, Superlinked made it incredibly practical.

Essential for anyone serious about search & RAG. Next cohort in March 👉 https://lnkd.in/gxnkckZa

Trey and Doug have put together and delivered an excellent course that includes both breadth and depth on the topic of AI Powered Search. I appreciated how they brought some dense topics to life through examples and interactive labs. There were plenty of opportunities to ask questions throughout the lectures and they offered weekly office hours to go in depth on topics of interest to the group. I recommend this course to any search practitioners that are looking to evolve their search products with leading edge techniques. I especially recommend this course to Search Product Managers who are new to the AI-Powered Search landscape that need a grounding in the technology.

After six months of intensive AI upskilling and building RAG systems, I thought I had a decent handle on retrieval . Then I took the course AI-Powered Search from Maven , and realized how much I'd been missing. Trey Grainger and Doug Turnbull go far beyond the "download embeddings from Hugging Face and hope for the best" approach. They systematically dismantle the naive assumption that hybrid search (BM25 + cosine similarity) is "good enough." As Trey put it in the semantic query understanding session: most people just throw embeddings and keywords at the search engine and hope it comes up with the right thing. This course is about challenging that paradigm.

Just wrapped up the Maven Course: AI-Powered Search - Modern Retrieval for Humans & Agents, and wow—it's a game-changer that left me buzzing with excitement and clarity!​I dove in initially for the product/search angle, but this course delivered extreme value for technical deep-dives and business insights, empowering owners to max out search in development. The content is razor-sharp up-to-date, unpacking complex solutions like personalized behavioral embeddings, agentic search paradigms, and ML ranking with crystal-clear intuition that ties every piece together—no more black-box confusion.​What blew me away? The perfect learning intensity, permanent access to recordings/notebooks/slides, and an amazing community of top-tier lecturers/peers —plus live 'office hours' for diving into any question, even off-scope. It's not just theory; it's hands-on mastery bridging demos to production systems that actually crush relevance for humans and agents.​If you're a dev, founder, or search-curious pro, this is your edge in the AI retrieval revolution—join!

I took this course because, even after doing search work for a couple of years (mostly self-taught), I kept running into the same problem: I could research concepts online, but the pieces didn’t always click into a production-grade mental model. This class was exactly what I needed. Doug and Trey bridge the gap between theory and real systems in a way that’s hard to get from blog posts and docs, especially the practical “how” behind query understanding, making hybrid retrieval behave predictably, and turning behavioral signals into an iteration loop that actually improves ranking. The biggest value for me was the experience transfer: the nuance, the “why,” and the decision-making that comes from having built and tuned search at scale. The live format and Q&A made a huge difference, you can bring your real problems and get grounded guidance. The hands-on labs were also excellent: a reusable set of notebooks and techniques I can apply to my own stack. If you’re building search/RAG systems and feel like you’re missing the foundations (or the production intuition), this course fills that gap fast.

Excited to share that I’ve completed Maven’s AI-Powered Search course .
Over 5 weeks, I went deep on what it takes to build modern, relevance-driven search systems—from classic IR foundations to LLM-era retrieval and ranking.
A few key takeaways I’m walking away with:
Relevance is a product problem as much as a modeling problem: user behavior insights, implicit signals, and evaluation loops matter as much as architecture.
Hybrid retrieval is the default: mixing sparse + dense (e.g., MiniCOIL-style approaches) and tuning the blend for real traffic.
Ranking is the real lever: LTR, click models, bias-aware learning, and active learning to keep improving without endless labeling.
RAG done right: interleaving strategies, retrieval quality, and grounding—because “more context” ≠ “better answers.”
Performance + UX: vector search optimization, semantic query understanding, semistructured embeddings, and agentic search patterns.
Big thanks to the speakers who shared practical lessons and real-world war stories Doug Turnbull and Trey Grainger🙏
Next up: applying these ideas to AI-powered conversational search—improving relevance, grounding, and measurable user outcomes.

I took this course because we hit a ceiling. We built a RAG-based search, and for a while, we were making progress. But at some point, we were just rearranging deck chairs, getting marginal improvements while the big problems weren't moving. Latency issues, gnarly quality gaps, the things that actually mattered. I realized I needed a deeper understanding of search and retrieval to make real gains. I had previously read Doug and Trey's AI-Powered Search book and decided to take their class. I'm also a fan of Doug's blog. What I found was a much deeper exploration of the topics in the book. They cover a lot of ground in the book, but the nuances come out in the class discussions and Q&A. Doug and Trey have worked in search and information retrieval for over a decade. When anyone in the class asked a question, they could pull from a variety of experiences to give a genuinely useful answer, not a theoretical one. You find yourself leveling up just from the conversation. Before taking the class, I felt like I had exhausted my options. Now I see there's a lot more room for improvement. They gave me new tools, but they also gave me perspective. Since they've worked on high-scale, top-tier search systems, they also set expectations for what excellence actually looks like. So you know what you're aiming for, not just what to try next. If you want to get good in this area, I can't think of a better class. I don't know of anyone else with this level of expertise who's actually teaching it.