The challenge
Hello Genie is a New Jersey based AI business assistant for small businesses, built for the bakery owner, the fitness coach and the freelance designer who should not be spending their evenings on data entry. Its published product covers four areas: smart scheduling, customer communication, document processing and workflow automation, connecting to Google Workspace, Microsoft 365, HubSpot, Salesforce, QuickBooks and Xero. It sells at 29, 79 and 199 dollars per month across Starter, Growth and Pro tiers.
The assistant does not stop at drafting. It reads untrusted inbound content, customer emails and uploaded invoices and contracts, and then acts on it: sending replies under the owner's name, booking meetings, and writing records into connected accounting and CRM systems. An unreviewed model output in that path is a sent email or a booked meeting, not a draft a human can quietly discard. That single fact shaped the whole program.
Their own published commitments set the bar. Hello Genie publishes 99 percent accuracy on data extraction, a number that has to survive a badly lit phone photo of a receipt. Communication that sounds like the individual owner rather than a generic professional register is the stated differentiator. And the economics have to work at 29 dollars a month, which makes model selection and prompt discipline a cost question, not just a quality one. Timing sharpened all of it: the engagement ran in July 2026, in the run up to the product going live, so the emphasis was release readiness rather than exploration.
What Kriv AI did
What runs in production today
Hello Genie runs Claude in production inside its shipped product, on Claude via Amazon Bedrock, with the team holding its own Anthropic organisation for internal work. A document extraction pipeline uses Claude's vision and multimodal capabilities: an owner photographs or uploads an invoice, receipt or contract, and Claude extracts line items, tax, totals and vendor details, matches them against chart of accounts templates and syncs the result to QuickBooks or Xero. A tone adaptive client assistant learns an owner's communication style from their previous email, reads context from calendar availability and service tiers, and drafts personalised replies and follow up sequences for one click approval. Both serve small business customers across the Starter, Growth and Pro tiers. This is a production deployment, not a pilot or a proof of concept.
Discovery first, then a calibrated curriculum
Week one settled what would be taught. Working sessions covered current and intended Claude usage, target workflows, output schemas, owner approval points and known failure modes across the four capability areas, alongside a review of customer supplied non sensitive prompt samples, schemas and sample documents. Three priority workflows were confirmed in writing as the teaching cases, recorded with the roster, schedule and success measures in a calibration memo the customer accepted.
Ten hours, five sessions, exercises inside the session
Delivered remotely to a cohort of five to ten named participants from product, engineering and operations, with hands on work inside session time rather than scheduled as separate labs.
- Claude fundamentals for the product team: the boundary between deterministic rules, existing product logic and model judgment, and which parts of the four capability areas warrant a model at all.
- Prompt engineering for customer facing communication: holding one individual owner's tone rather than a generic professional register, follow up sequences, priority flagging, and preventing invented availability or unsupported commitments in mail sent under the owner's name.
- Document processing and structured outputs: extraction from invoices, receipts, contracts and forms across PDF, image, Word and spreadsheet inputs, schema conformant JSON for downstream accounting and CRM destinations, poor scans and edge case layouts, and field level validation so low confidence fields surface for the owner rather than filing silently.
- Tool use and workflow automation: tool and function definitions, multi step orchestration with conditional logic across calendar, email, CRM and accounting connectors, idempotency, retries and error handling, and approval steps, audit trails and safe rollback so automated actions stay reversible and under owner control.
- Evaluation, safe deployment and cost control: test sets from de-identified cases, scoring rubrics calibrated against the extraction accuracy Hello Genie publishes, regression testing, prompt injection handling for adversarial content in inbound email and uploaded documents, and model selection and prompt caching against the monthly price tiers.
Reversibility as a design principle, not a feature
Because every automated action is real, the program treated owner control as structural. Approval steps, audit trails and safe rollback were established as patterns for every action the assistant takes, and low confidence extraction was designed to surface for attention rather than file silently. The point was not to make the model more confident, it was to make the system honest about when it is not.
No customer data, written into the contract
Kriv AI did not request, receive, access, store or process personal information relating to Hello Genie users or their customers, mailbox or calendar contents, real invoices, contracts or business records identifying real people or organizations, or production credentials, and held no access to production systems, data stores or connected third party accounts at any point. Every teaching example ran on synthetic or de-identified data the customer supplied or approved, and all exercise work was executed by Hello Genie participants inside their own environment. The engagement also explicitly excluded connector development with third party providers, security certification and penetration testing, and any determination of the accuracy or sufficiency of an extraction, scheduling decision or customer communication produced with Claude assistance. Those remained Hello Genie's responsibility and the contract said so.
Results
- 99.4 per cent extraction accuracy on unseen receipts and handwritten invoices, with a median 4 seconds to extract and structure a line item invoice.
- 88 per cent of processed documents clear the automated confidence threshold without owner review, saving 7.5 hours a week per business owner on routine bookkeeping.
- Average response latency to a new client inquiry fell to 1.5 minutes, down from 4 hours, with 91 per cent of drafted replies sent by owners without any edits.
- A 32 per cent increase in client booking conversion within 30 days of setup, and more than 50 pre built automation workflows launched from the prompt library.
- Hello Genie uses Claude in day to day work operations, on its own Anthropic organization, against the workflows established during the program.
- All ten instruction hours delivered on schedule across three consecutive weeks.
- A cohort of product, engineering and operations staff able to design, test and iterate Claude prompts and workflows across scheduling, customer communication, document processing and automation without external assistance.
- All three priority workflows documented with working prompts and stated limitations in a repo ready reference pack, accepted against written criteria.
- A repeatable evaluation method handed over, with test sets, scoring criteria and pass and fail thresholds, so extraction and communication quality can be measured before release and re-measured after a change.
- Documented internal standards for data minimization in prompts and logs, untrusted input and prompt injection handling, and owner approval points before the assistant acts.
- A prioritized 90 day roadmap mapping Claude capabilities to workflows with effort and dependency notes, aligned to what each pricing tier gates.
About Hello Genie
Hello Genie is a New Jersey based AI business assistant for small businesses, covering smart scheduling, customer communication, document processing and workflow automation, with connections to common business tools including Google Workspace, Microsoft 365, HubSpot, Salesforce, QuickBooks and Xero. For more information, visit hello-genie.com.
Disclosure
Figures in this story are as reported by Hello Genie. Kriv AI did not independently audit Hello Genie's systems or records, and held no access to production data at any point.