Revolut's Real-World Lessons in Deploying GenAI
nathanbenaich · x · 2026-07-19
Revolut's Nikolay Donets shared practical insights from their large-scale AI rollout:
- The company now has twice as many GenAI use cases as traditional ML, spanning 40+ countries and 200+ products.
- They encountered deployment issues like "a model in the fallback chain silently failing, leaving the team completely blind."
- Generative model costs are also substantial, with frontier models costing up to 8x more than alternative solutions.
- At a scale of 25,000 calls per month, their voice assistant has increased issue resolution speed by 8x.
More from coding & agent
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- How I let my cloud Grok Bot code on my Mac mini via Tailscale, no public ports — talkaboutdesign · 2026-10-27
- Context layout is a cost decision: one memory edit can bust your entire agent prompt cache — dscape · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy makes all ~2,850 AWS bindings work from any cloud, not just S3/Lambda/DynamoDB — samgoodwin89 · 2026-10-27