Senior Site Reliability Engineer

Tripadvisor demos AI planning tool. Tripadvisor recently partnered with Nvidia, Nebius and Nexla to show what the future of end-to-end travel planning could look like with artificial intelligence (AI). The company is looking for ways to streamline travel planning - to make it simpler, smarter and more personalized, Rahul Todkar, vice president and head of data and AI for Tripadvisor, wrote on LinkedIn. In a live demonstration at Nvidia GTC 2026, attendees inserted an influencer's video into an experimental tool on Tripadvisor's native AI experience, he wrote. The tool found locations and experiences featured in the video, cross-referenced them with Tripadvisor's data and created a personalized itinerary that was "ready-to-book" for the user, according to Todkar. "Planning a trip is exciting...but it's also complex," Todkar wrote, sharing a demo video. "Hours (sometimes weeks) go into researching and sorting through ideas, options and prices across dozens of sources. As AI and agentic workflows evolve, our teams have been experimenting, testing and iterating on ways to make that process easier." Todkar told PhocusWire Tuesday that Tripadvisor offered the travel focus, Nvidia served as the model provider layer, Nibius provided the backbone and Nexla offered data integration. "Knowledge and expertise is what really is the key here." While Todkar said it wasn't that hard to pull the feature together, he did identify a couple challenges. Video processing, for example, proved to be difficult. "Understanding the right context, meaning out of the videos and then mapping that to our existing POIs [points of interest]... all the things we have on our site... connecting those two dots was the number one challenge," he said. Personalization was another hurdle, according to Todkar, who said the company has been performing specific experiments to learn more about both. Testing and experimentation comes next. "We do explore and do a lot of pilots on our site, especially with our AI-focused chip planning application," Todkar said. "This will be one of those features. We'll try to test and see what the user response is." If all goes well, the feature can be pushed into live production. Last year, Expedia Group debuted its AI-powered Trip Matching feature, which allows users to turn Instagram Reels into bookable itineraries. When asked about the similarity between the two initiatives, Todkar said he's seen early versions of the feature. He said the concept is the same but the execution might be a little different. "Inspiration can come from anywhere," he said. "More and more inspiration is your social media, your AI platforms, your top of the funnel. What users are increasingly telling us is they want to take that inspiration into trip planning, then booking." Inspiration to planning to booking has to become seamless. "All brands, I'm sure, are thinking about that," he said. "The differentiation becomes, how do you bring your own data? How do you bring your own personalization? And how do you drive actions faster?" Todkar said Tripadvisor is pushing further into AI-first features. "This is one of those features," he said. "There's a lot more we have in the hopper, and this is just an exciting space to be in."

Nexla, an AI-powered data integration platform, has partnered with Vespa.ai to simplify real-time AI search across enterprise data sources. The collaboration addresses a critical challenge in AI application development: connecting and preparing data from hundreds of disparate sources before powering intelligent search systems. Nexla offers over 500 pre-built connectors that transform data from various enterprise systems into production-ready formats, whilst Vespa provides distributed search, vector retrieval and real-time inference capabilities. The partnership includes two new integrations: a Vespa Connector in Nexla for seamless data piping, and a Vespa Nexla Plugin CLI that automatically generates application packages. The solution targets organisations building AI search, retrieval-augmented generation applications and high-throughput systems serving billions of documents with real-time updates.

Nexla and vespa.ai partner to simplify real-time AI search across hundreds of enterprise data sources. By GlobeNewswire February 18, 2026 Native integrations reduce setup time and ongoing maintenance by making it easy to ingest, index, and continuously update data from enterprise systems SAN MATEO, Calif., Feb. 18, 2026 (GLOBE NEWSWIRE) - Nexla, the enterprise-grade AI-powered data integration platform for agents, today announced a strategic partnership with Vespa.ai, the creator of the leading AI search platform for building and deploying large-scale, real-time AI applications. The partnership eliminates one of the biggest bottlenecks in AI application development: getting production-ready data into scalable, high-performance AI search and retrieval systems. Organizations building AI-powered applications face a critical challenge: connecting and preparing enterprise data from hundreds of disparate sources before it can power intelligent search and retrieval. This data variety - structured/unstructured, batch/ streaming, modern/ legacy, creates complexity that slows AI deployment to production. With over 500 pre-built connectors, Nexla addresses this challenge by transforming data variety from any enterprise system into production-ready data products for AI and agents, while Vespa provides the distributed search, vector retrieval, and real-time inference capabilities required to serve AI-powered applications at scale. Together, they create a seamless path from raw enterprise data to intelligent, production-grade AI search. As part of the partnership, Nexla launched native Vespa integrations that make working with Vespa faster and simpler: * Vespa Connector in Nexla: Seamlessly pipes data from sources such as Amazon S3, PostgreSQL, Snowflake, APIs, and even existing vector databases directly into Vespa, without custom code or complex configurations. * Vespa Nexla Plugin CLI: Automatically generates draft Vespa application packages, including schema files, directly from Nexla's metadata-defined data products (Nexsets), dramatically reducing setup time and configuration errors. Advertisement These capabilities enable teams to migrate from other vector databases, sync operational databases into Vespa, or continuously update Vespa indexes using batch, streaming, or CDC pipelines, all without writing code. The combined solution is especially valuable for organizations building or scaling: * AI search and RAG applications requiring hybrid retrieval across vectors, keywords, and structured filters * High-throughput, low-latency systems serving billions of documents with real-time updates * Complex ranking and inference pipelines, including multi-phase ranking and LLM integration Advertisement Nexla prepares and governs the data; Vespa executes advanced retrieval, ranking, and inference where the data lives. "Data integration and intelligent retrieval are two sides of the same coin in modern AI architectures," said Saket Saurabh, CEO and Co-Founder of Nexla. "Nexla unlocks data variety, transforms it, and delivers enterprise-grade, ready-to-use data products; Vespa.ai makes that data searchable and actionable in real time. This partnership creates a powerful combination for organizations building agentic RAG, recommendation systems, and AI-powered search at scale. Together, we're removing the friction between data preparation and intelligent retrieval, so teams can focus on building transformative AI experiences instead of wrestling with data plumbing." "Vespa is built for teams that need precision, performance, and real-time control at scale," said Jon Bratseth, CEO of Vespa.ai. "By partnering with Nexla, we're removing friction between data preparation and real-time execution, so teams can move from raw enterprise data to production-grade AI search and RAG systems faster and with far more control." https://www.nexla.com/nexla-vespa-ai About Nexla Nexla is an enterprise-grade, AI-powered data integration platform for agents that unlocks data from any source and transforms it into production-ready data products for AI and agents. With support for 500+ pre-built connectors and multiple integration styles - including ELT, ETL, streaming, APIs, and agentic RAG. Nexla enables teams to build and manage data flows without writing code. Trusted by leading enterprises, Nexla processes over one trillion records per month across industries. Learn more at nexla.com. About Vespa.ai Vespa.ai is a powerful platform for developing real-time search-based AI applications. Once built, these applications are deployed through Vespa's large-scale, distributed architecture, which efficiently manages data, inference, and logic for applications handling massive datasets and high concurrent query rates. Vespa delivers all the building blocks of an AI application, including vector database, hybrid search, retrieval augmented generation (RAG), natural language processing (NLP), machine learning, and support for large language models (LLM) and vision language models (VLM). It is available as a managed service and open source. Learn more at vespa.ai. Media Contact Jayashree Rajan [email protected]