Proprietary APIs offered zero control. Moving to managed open models secured 2-second answers and fixed pricing for news search.
A major American news publisher serving millions of users with award-winning journalism and deeply sourced reporting.
The organization wanted to launch a public-facing Q&A tool but found proprietary APIs offered zero control and unpredictable costs. Self-hosting...
“Together AI's flexibility in creating customized configurations for our journalism use case has been invaluable to our success.”
Major daily newspaper and digital media organization for global news.
AI-native cloud platform for training, fine-tuning, and deploying open-source models.
The Washington Post's News content search is part of this use case:
Related implementations across industries and use cases
Researching millions of articles took 30 minutes. An AI tool now retrieves facts, summaries, and angles in seconds.
Fragmented pipelines slowed cross-site suggestions. A unified AI vector database cut latency 90%, processing 1,500 queries per second.
Teams manually read every new comic to extract keywords. Now, AI analyzes art and speech bubbles to auto-generate campaign metadata.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Fragmented vendor feeds obscured matching items. Now, AI pipelines parse millions of unstructured records into curated collections.
The CPO sourced every candidate by hand, plus a full team. AI now builds targeted pipelines in 24 hours—she runs it solo.
Asset ledgers across three tools were unreachable by the AI helpdesk. One migration consolidated them — now employees self-serve through AI.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.
Proprietary APIs offered zero control. Moving to managed open models secured 2-second answers and fixed pricing for news search.
A major American news publisher serving millions of users with award-winning journalism and deeply sourced reporting.
The organization wanted to launch a public-facing Q&A tool but found proprietary APIs offered zero control and unpredictable costs. Self-hosting...
“Together AI's flexibility in creating customized configurations for our journalism use case has been invaluable to our success.”
Major daily newspaper and digital media organization for global news.
AI-native cloud platform for training, fine-tuning, and deploying open-source models.
The Washington Post's News content search is part of this use case:
Related implementations across industries and use cases
Researching millions of articles took 30 minutes. An AI tool now retrieves facts, summaries, and angles in seconds.
Fragmented pipelines slowed cross-site suggestions. A unified AI vector database cut latency 90%, processing 1,500 queries per second.
Teams manually read every new comic to extract keywords. Now, AI analyzes art and speech bubbles to auto-generate campaign metadata.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Fragmented vendor feeds obscured matching items. Now, AI pipelines parse millions of unstructured records into curated collections.
The CPO sourced every candidate by hand, plus a full team. AI now builds targeted pipelines in 24 hours—she runs it solo.
Asset ledgers across three tools were unreachable by the AI helpdesk. One migration consolidated them — now employees self-serve through AI.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.