Dec 12, 2023 · 28m · big-technology
How Generative AI Is Changing Travel — With Priceline's Marty Brodbeck
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Priceline Chief Technology Officer Marty Brodbeck discusses the company's practical implementation of generative AI across conversational trip planning, checkout optimization, customer support, and multi-cloud architecture.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Alex holds 30.6% of the talking time here. How this is scored →
Alex as informed peer 3.9 Guest teaching 4.8 Guest disagreement 0.0 Alex pushing back 0.4
speaking balance: gold is Alex, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 19:42 Tempered assessment of AI code generation
Marty pushes back on the industry narrative that AI coding tools are an unqualified success, noting mixed pilot results and developer hesitancy to commit AI code to production.
Hardest push from Alex ▶ 19:00 Host challenges AI developer productivity claimsAlex puts Marty on the spot, noting that tech leaders often call coding assistants magical but fail to provide measurable productivity numbers.
Biggest teaching moment ▶ 12:23 Reframing LLM memory around enterprise CDP infrastructureMarty educates Alex on data architecture, explaining that persistent user memory is handled by proprietary customer data platforms rather than the LLM itself.
Alex holds their own ▶ 19:00 Alex cites reporting on engineering managementAlex leverages his reporting background covering tech teams to question whether generative AI tools are yielding concrete, measurable productivity gains in software engineering.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Alex as informed peer | Guest teaching | Guest disagreement | Alex pushing back | Why |
|---|---|---|---|---|---|---|
| Integrating AI into Checkout and Post-Booking Flows | 3 | 5 | 0 | 1 | Alex poses the open question of whether users actually want a conversational bot versus a traditional GUI. Marty outlines Priceline's four main Generative AI pillars and details their AB-testing methodology for integrating LLMs directly into checkout. | |
| Developing the Conversational AI Travel Concierge | 4 | 4 | 0 | 0 | Alex offers his personal workflow for filtering hotels by location and Wi-Fi quality. Marty connects this to conversational search capabilities and real-time user query trends like pet-friendly policies. | |
| Mitigating LLM Hallucinations and Managing System Costs | 4 | 6 | 0 | 0 | Alex asks about discovering deep-tail data through AI search. Marty details how marrying proprietary data with external mapping APIs prevents hallucinations, creates authoritative cached stores, and keeps computational costs manageable. | |
| Leveraging Customer Data Platforms for Persistent Memory | 4 | 6 | 0 | 0 | Alex raises the issue of LLM short-term memory constraints in travel planning. Marty clarifies that long-term contextual memory is solved by underlying customer data platforms (CDPs) rather than the LLM itself. | |
| Checkout Strategy and AI-Generated Marketing at Scale | 4 | 5 | 0 | 1 | Alex questions why Priceline prioritized checkout over top-of-funnel discovery, then inquires about marketing content generation at scale. Marty explains checkout friction points and their multi-tier hallucination check pipeline. | |
| Operational Support with Penny and Developer Productivity | 5 | 6 | 0 | 1 | Alex challenges the hype around AI coding assistants by putting Marty on the spot regarding concrete productivity metrics. Marty provides a candid assessment of pilot results, citing developer reluctance to commit AI code to production. | |
| Cloud Ecosystem Innovation and True One-to-One Personalization | 5 | 5 | 0 | 0 | Alex reflects on multi-vendor cloud architectures and non-traditional audience segmentation. Marty describes moving past static demographic buckets toward dynamic, session-based one-to-one personalization. | |
| Concluding Insights on Generative AI in Travel | 2 | 1 | 0 | 0 | Alex wraps up the conversation, thanking Marty for detailing practical enterprise production deployments of generative AI in travel. |
Statements from this episode (12)
Priceline Rolls Out Generative AI Checkout Bot to 100% of Users
“We ran tests to make sure that the bot had a positive impact on the user experience and the, you know, and revenue, and after several iterations of prompt engineering, we got to a point where the bot was A-B testing positive, and so we pushed it out to a hundr…”
Priceline Develops Natural Language Free-Text Travel Concierge Features
“Now we shifted our focus towards what we're calling The travel concierge part of our business, which is really the trip planning, and we're currently working on capabilities to allow users to do free tests, free text searching.”
Grounding LLMs With Proprietary Data Reduces Hallucinations and Improves Caching
“Marrying the large language model with our own data sets with third-party data really drives down, ah, the amount of hallucinations that we get from these kinds of conversation and almost creates authoritative sources of cash information around hotel reviews, …”
Unmanaged LLM API Calls Cause Generative AI Costs to Skyrocket, Brodbeck Warns
“Cost in, in running and managing a generative AI platform could skyrocket. So you have to be very smart about when you refresh or make direct calls to the large language model and this third party information.”
Brodbeck: AI Personalization Depends on Underlying Data Architecture, Not LLMs
“Well, I think this is more about how companies design and develop their data architectures than generative AI understanding personalization on the individual level.”
Priceline Connects Customer Data Platform Directly to Generative AI Interface
“And so what we've done is we've married all that rich CDP data so that when you log into the gender of AI experience, a signal is fired that understands who you are. So in this case, you're Alex and I understand what you searched on, what you've booked in the …”
Priceline Generates AI Marketing Content for Top 1,000 Global Destinations
“So we've taken the top 1000 city destinations in the world and have been systematically creating content in those three different channels I talked about.”
Priceline Validates AI Outputs Against Google Places API to Prevent Hallucinations
“We've built a hallucination service that verifies the information using third-party information sources. Like in the case for a lot of our hotel products, we're using Places API from Google. Which is really an authoritative source for a whole host of geographi…”
Priceline Sees Significant Savings From Post-Booking Generative AI Support Chatbot
“I can't share the exact results, but we're seeing significant operational efficiency and savings By using this technology as part of our post-booking flow and have plans to expand it to all of our product lines in the foreseeable future.”
Priceline Saw Mixed Results in Generative AI Code Automation Pilot
“So we've just recently completed a pilot of a product that played in the generative AI space as it pertains to code automation. And quite frankly, we had mixed results.”
Generative AI's True Developer Value Is Automating Maintenance Tasks, Says Brodbeck
“And that's where I see the real value in generative AI is unlocking a lot of the maintenance and support functions that developers have to do and having them solely focus on Building great features for your product.”
Static Audience Tiers Fail Because Customer Travel Contexts Constantly Change
“What we're really striving for is really one-to-one personalized, a one-to-one personalized product experience, because The same person that you could put into an audience the next time they come back to Priceline could be looking for something completely diff…”