Trainline is Europe’s largest train booking platform serving over 27 million customers. Together, we’re building voice agents that automate refunds when trains are cancelled or delayed. Our agents integrate directly with @thetrainline
's telephony and ticketing systems – and can
AGENTS
-
ElevenLabs builds voice agents for Trainline refunds automation
By
–
-
God of Prompt shares real AI prompts and frameworks for business
By
–
This is God of Prompt. New home, same mission.
I share the prompts and systems I actually use to run business, marketing, build agents, and ship real work with AI. No "10 tools" threads.
Real frameworks, with receipts. What you get by following:
– New guides and prompts every -
Loop engineering: design loops that prompt agents; unattended loops fail.
By
–
about loop engineering. everyone's saying the same thing this week. you don't prompt agents anymore, you design loops that prompt them. here's the job that loop hands right back to you. a loop running unattended is also a loop failing unattended. loop engineering takes you
-
Spans show outputs not intent; logging reasoning helps
By
–
Right. Spans show what got called, not what the agent thought it was doing in between. Ollie walks the causal chain to find where it broke, but intent still gets pieced together from outputs. Logging the model's reasoning as its own span helps, but most frameworks skip it by
-
Google Search gets biggest upgrade in 25 years with AI intent-based experience
By
–
The search box is getting its biggest upgrade in 25 years. @Google is moving from keyword matching to a fully agentic, intent-based experience powered by @GeminiApp.
— Harold Sinnott 📱 (@HaroldSinnott) 9 juin 2026
Search isn't just about finding links anymore; it's about building custom solutions. #AI
Check out the full… https://t.co/XgghNyUf1zThe search box is getting its biggest upgrade in 25 years. @Google is moving from keyword matching to a fully agentic, intent-based experience powered by @GeminiApp
. Search isn't just about finding links anymore; it's about building custom solutions. #AI Check out the full -
Multi-Agent Lite Swarms: Opus, GPT, Deepseek, Gemma, 10x Cheaper
By
–
🚨 Multi-Agent – Lite Agent Swarms – Optimize Cost On Large Agentic Loops
— Abacus.AI (@abacusai) 8 juin 2026
After a lot of experimentation we have open-source AI agent swarms live!!
– Opus 4.8 and GPT 5.5 do the planning
– Deepseek flash and Gemma do the work
– Perfect for multiple parallel tasks
– 10x cheaper… pic.twitter.com/ZzKiyiFl7JMulti-Agent – Lite Agent Swarms – Optimize Cost On Large Agentic Loops After a lot of experimentation we have open-source AI agent swarms live!! – Opus 4.8 and GPT 5.5 do the planning
– Deepseek flash and Gemma do the work
– Perfect for multiple parallel tasks
– 10x cheaper -
Gratitude for focus on real problems over wrappers and agents
By
–
I am glad more and more people are starting to tackle the actual real problems rather than building wrappers and “agents”
-
Agent loops need task repetition, code execution, and rejection gates
By
–
Agent loops are not a default, even if it's viral.. They only work when the task repeats, the agent can run the code, and something inside the loop can reject bad output. Miss that gate and you are just automating token burn.
-
New Release: Machine Learning Engineering on AWS Book
By
–
New Release from @PacktPublishing @PacktDataML "Machine Learning Engineering on AWS: Build, deploy, and operationalize LLMs, AI agents, and generative AI systems on AWS" — at https://
amzn.to/4od7C1I 𝗪𝗵𝗮𝘁 𝗬𝗼𝘂 𝗪𝗶𝗹𝗹 𝗟𝗲𝗮𝗿𝗻:
Build and deploy AI agents using -
METR report: most SWEBench results unmergeable; FrontierCode unsolved by frontier models
By
–
It's finally out!!! @METR_Evals found that more than half of SWEBench results is unmergeable slop. FrontierCode represents over 1000+ hours of maintainer validated software engineering work most frontier models cannot yet solve, much less solve with high quality. Cog had IOI