AI & Automation
AI that lives inside your business, not beside it.
Voice agents, document readers, forecasts and checking agents, built on your data, in your accounts, with a person in the loop — by the team that also builds the ERP, LMS and website they connect to.
- 789 ms
- ~$0.03/min
- 40% faster audits
- $15K/yr saved
- 451 marking schemes
Four places AI pays for itself.
Calls and enquiries
An agent that answers, qualifies and books, in English or Urdu.
- Phone, web and WhatsApp on one brain
- Books the meeting and confirms it mid-call
- Writes up every call
789 ms, ~$0.03/min.
Learn more about Calls and enquiriesPaperwork
Invoices, receipts and reports read into your accounts or ERP, on your own machines if you want.
- Posts to Sage and Microsoft Dynamics
- A second model checks every field
- Reports turned into summaries and Urdu audio
4 of 5 receipts perfect; $15K/yr saved.
Learn more about PaperworkStock and demand
Forecasts by product, store and day that a planner can adjust and lock.
- Three model families compared on your history
- Data pipelines with health checks
- Product data cleaned at millions of records
3 model families, SKU × store × day.
Learn more about Stock and demandChecking and monitoring
Agents that audit, scan regulators, mark papers and search private documents, and say "I don't know" when they should.
- Audit and decision routing
- Daily regulator scans, on premise
- Marking against the official scheme
Audit decisions 4 h → 5 min.
Learn more about Checking and monitoring
Ava · our voice agent
Hear Ava take a call.
[TODO] Recorded inbound and outbound calls, with transcripts, go here once cleared for public use.
789ms
median reply
821ms
95th percentile
~$0.03
per call-minute
Measured on our own server across logged calls, Oct 2026. Carrier and AI billed at cost to your account; no platform markup.
What happens when she's unsure
- Says she's an AI on every call
- Transfers to a person when confidence drops
- Checks the do-not-call list before every dial
- Your data is never used to train models; you hold the API keys; on-premise available
| Platform | Per call-minute (Oct 2026) | Notes |
|---|---|---|
| Ava (Web Wroughts) | ~$0.03 | Carrier + AI at cost; you hold the keys |
| Retell AI | $0.07–0.31 (typical ~$0.11–0.13) | Add-ons per minute; billed per second incl. silence |
| Vapi | ~$0.08–0.17 all-in | $0.05 platform + STT/LLM/TTS/telephony |
| Bland AI | $0.11–0.14 + $299–499/mo plans | Transfer fees $0.04–0.05/min |
| Air.ai | $0.20–0.40; ~950 ms median latency | Monthly plans from ~$1,499 |
Four things we put in writing about AI.
Measured, not claimed
Latency, accuracy and cost figures come from our logs and are dated.
How we use AI: Measured, not claimedUsage at cost
Carrier minutes and AI tokens are billed to your own accounts at the provider's price. No platform markup.
How we use AI: Usage at costYour data stays yours
Never used to train models; you hold the API keys; on-premise on your own GPUs is an option.
How we use AI: Your data stays yoursA person in the loop
Every agent says it is an AI, hands over to a human when unsure, and never takes a high-stakes action (payment, database write, deletion) without an approval gate you set.
How we use AI: A person in the loop
How we keep it honest.
Grounded answers
Every reply cites your documents, and the agent abstains when it doesn't know.
Case 29: MRR 0.900, 5 of 5 correct abstentions.
Human handoff
A warm transfer to a person, and a do-not-call check before every dial.
Your keys, your data
You hold the API keys, and your data is never used for training.
On-premise option
Open models on your own GPUs, so documents never leave the building.
Cases 12, 17, 20.
Evaluation before go-live
Labelled test sets, automated evals, staged UAT, and approval gates on high-stakes actions.
Where it plugs in.
- 01Website / CRM
- 02Voice / chat agent
- 03ERP / LMS
- 04Document AI
- 05Forecasting
We build every box.
AI we've built.
- Voice AI
Ava answers a lead in under a second
789 ms median reply time, measured · our product · Sep to Oct 2026
- Banking
Audit agents for 1,200+ bank users
40% faster audit cycles · HBL
- Finance software
Invoice and receipt reading on local GPUs
$15K a year saved on cloud AI ·
- Document AI
Receipt Reader: nothing leaves the machine
4 of 5 real receipts read perfectly; the fifth off by 1 rupee · receipts and invoices · 2025 to 2026
- Voice AI
One voice agent on phone, web and WhatsApp
3 channels · services company in Canada · 2026
- EdTech AI
Marking handwritten exam papers with AI
451 marking-scheme PDFs registered · prototypes by Anees and the Zynthrix team · Mar to Apr 2026
- Retail AI
Demand forecasting a buyer can question
3 forecasting model families compared · fashion and FMCG · 2026
- Document AI
Briefcast: company reports as Urdu audio briefings
~186 documents processed on the production server · 2026
- Compliance
Daily regulatory compliance, on premise
Daily scan of regulator updates · on-premise deployment
- Private search
Search that says "I don't know"
1,169 video transcripts · lab · Sep 2026
How it's priced.
- Every AI build starts with a 5-Day Proof Sprint ($490, pay on approval): a working prototype on your data.
- Then a fixed build price (single-flow agent typically $1,500–6,000; multi-system agents scoped per module) or the Software + AI team at $3,200/mo.
- Carrier minutes and AI tokens billed at cost to your accounts; tuning from $500/mo.
Prices exclude taxes and third-party costs (hosting, carrier minutes, AI usage), billed at cost to your own accounts.
See all pricesQuestions about AI.
Will it hallucinate, and what happens when it does?
It can, so we build for it. Answers come from your own documents and cite them. When the agent can't find support, it says it doesn't know and hands over to a person: our private search abstained correctly 5 times out of 5 in testing. Payments, database writes and deletions always wait for an approval you set.
What does a call or a document cost?
Carrier minutes and AI tokens are billed by the providers to your own accounts at their published rates. We add no platform fee. Typical: ~$0.03 per call-minute on Ava; cents per document for reading; near zero for on-premise.
Can it run on our servers?
Yes. Open models can run on your own GPUs, so nothing leaves the building. We've done this for invoice reading, daily compliance checks and a receipt reader.
Which languages?
Our voice agents speak English and Urdu today, and our invoice reader handles English, French and Spanish. Tell us yours and we'll test it in a Proof Sprint before you commit.
Does it replace staff?
It takes the repetitive part: first replies, data entry, routine checks. Your staff handle what it hands over, and they sign off each part before it goes live.
Who is liable when it's wrong?
Your contract sets liability, under English law. We also design so mistakes stay small: the agent says it's an AI, hands over when it's unsure, and can't make a payment, write to a database or delete anything without your approval.
How do you test it before it goes live?
On labelled test sets built from your own data, with automated evaluations, then staged UAT with the people who will use it. High-stakes actions stay behind approval gates after launch.
Send us one job.
Tell us what you need built. You'll have scope, price and dates in writing today, the work in 3–5 business days, and you pay only if you approve it.
Prices exclude taxes and third-party costs (hosting, carrier minutes, AI usage), billed at cost to your own accounts.