Get tasks doneBusiness size: Company

ZUS Coffee used live order lookups and human handoffs across support channels

Ruby · Malaysia · Filum.ai AI Agent, Filum.ai Voice of Customer

What the agent did

Filum queries live order records, answers routine questions and prepares ticket tags and summaries.

Result, as claimed

Filum reports 55%+ fully resolved without human handoff in Malaysia and Singapore in the first two months, and ~87% goal achievement to the team's criteria.

This is the result as claimed in the public source linked below. It has not been checked and is not a typical outcome.

Proof

Read the original public source

Try it yourself

Test when an order-status reply should go to a person.

Difficulty: Beginner

Prompt

From [synthetic chats], [mock order rows] and [approved policy], draft an answer only when the order ID and status match. For missing, conflicting or disputed records, make a short handoff note. List the policy line used. Keep all outputs in review and do not send.

Prompt: starter by Solopreneur Code

Tools

Steps

  1. Create mock orders and include a wrong-account test.
  2. Mark which cases a person should handle.
  3. Compare drafts to the approved policy and inspect false resolutions.
  4. Have a person sign off on the trial. Stop before sending or connecting live data.

Watch out: A plausible order ID can expose someone else's status. Keep samples synthetic and follow applicable privacy duties, including Singapore PDPA where relevant.

Starter version by Solopreneur Code

Sign up for the newsletter

Get the next story from Solopreneur Code in your inbox.

Form not loading? Subscribe on Substack

Back to all stories

More like this

Browse all stories

Once a week it reads his spam and archived email, groups it by sender and looks for events and articles that fit his goals for the year. For each useful sender it gives up to 5 action points and suggests pulling it out of spam or unsubscribing. It now checks his calendar and working hours before it shortlists events. Claude also names new files and chats to one convention and drafts unusual workshop formats for him to pick from.

Result, as claimed

The first run went through 718 emails and it now handles about 30 a week. He says AI gives him back 3 to 4 hours a week, and one webinar it flagged put him on a path to new work.

Get tasks doneBusiness size: SoloIncludes prompt

A marketing educator has Claude route her inbox into CRM deals, tasks and draft replies, and says agents save her 3 to 4 hours a week

Cristina Peñaloza Morales · Country unknown · Claude, Gmail, HubSpot, Todoist, Notion, Riverside, Gamma

Reads new email and routes each one. Training enquiries become a HubSpot contact and deal with a drafted reply. Community messages get a draft reply. Possible collaborations become a Todoist task. Invoices and admin get a label. Also runs a scheduled Monday AI news digest.

Result, as claimed

She says agents save her between 3 and 4 hours a week.

Get tasks doneBusiness size: SoloIncludes prompt

A newsletter writer's Claude Code agent turned screenshots into 38 scheduled Substack Notes he did not write, with no engagement lift

Dheeraj Sharma @genaiunplugged · Country unknown · Claude Code, SubflowAI Substack connector, macOS scheduler, Substack Notes

Each morning a headless Claude Code run takes the oldest screenshot, classifies it, writes a 45 to 115 word Note in his voice, checks it against copy gates, uploads the image, schedules it for the next free day, reads the schedule back and archives the screenshot. Screenshots with emails, phone numbers or orders are set aside.

Result, as claimed

29 Notes scheduled by 23 Sep and 38 by 12 Oct with none written by him. 12 measured Notes had a median of 3.5 likes against 5 for his other Notes, which he reads as no lift.