Skip to content
AtheronLABS

You're visiting from the United States. Prices are shown in US dollars. Not right?

labs@atheron:~/industries/professional-services/ai$ train agent --grounded

Knowledge search

Your past work, found in seconds.

Most firms have already written the answer somewhere. A knowledge assistant over your own documents finds it, quotes it and links to the source, and keeps your clients' documents inside your own cloud account.

// what matters here

What is different in this industry.

01

Why an open-source model here

Client documents are confidential, so the model runs in your own cloud account and nothing reaches an outside model provider.

02

Answers with sources

Every answer shows the passages it used and links to the documents, so staff can check it.

03

Your permissions, kept

People only find what their existing folder permissions already allow them to open.

04

No professional advice

The assistant finds your own past work. It gives no accounting, tax or engineering advice of its own.

// example projects

Priced examples for this service.

Each one is hypothetical, labelled as such, and priced live from our rate card. Open any of them in the estimator and make it yours.

Example projectWeb appAIOpen-source model

Knowledge assistant over past work

Hypothetical. Not a client, not a result.

// The problem

A consultancy's staff spend hours finding past proposals, reports and methods in shared drives, and new joiners never find them at all.

// What we would build

A search and question tool over the firm's own documents, which answers with the passages it found and links to the sources, so staff can reuse past work. It runs on an open-source model in the firm's own cloud account, because client documents should not reach an outside model provider, and it gives no professional advice of its own.

// What is in it

  • Search and questions over your own documents
  • Every answer with the passages and links it used
  • Access follows your existing folder permissions
  • No professional advice, only your own past work
  • Runs in your own cloud account

// Stack

  • Open-source model
  • Next.js
  • PostgreSQL with pgvector
  • Document store interface
  • AWS

// estimate

Build
≈ US$96,600 to US$148,000, delivered within 22 weeksCAD 137,600 to 210,400
Hosting
≈ US$7,020 a monthCAD 10,000 a month
Support
≈ US$2,500 a monthCAD 3,565 a month

Prices in your currency are estimates from today's Bank of Canada rate. All invoicing is in CAD or USD.

AI route
Open-source model
Model running cost
≈ US$6,530 a monthCAD 9,300 a month

Timeline by milestone

Discovery
2.9 to 3.4 weeks
Specification and evaluation plan
0.4 weeks
Design approved
2.4 to 3.4 weeks
Core features
1.4 to 1.9 weeks
Full build
0.9 to 1.4 weeks
Working pilot
3.9 to 6.9 weeks
Testing and fixes
0.9 to 1.9 weeks
Launch
0.9 to 1.4 weeks
Production
0.9 to 1.4 weeks

Outside our hands, and added to the calendar

Your firm's security review
4 to 8 weeks

How it is paid

Deposit 20%
CAD 27,460 to 42,020
Discovery 1.8%
CAD 2,471.40 to 3,781.80
Specification and evaluation plan 10.5%
CAD 14,416.50 to 22,060.50
Design approved 1.8%
CAD 2,471.40 to 3,781.80
Core features 5.6%
CAD 7,688.80 to 11,765.60
Full build 3.7%
CAD 5,080.10 to 7,773.70
Working pilot 31.7%
CAD 43,524.10 to 66,601.70
Testing and fixes 1.8%
CAD 2,471.40 to 3,781.80
Launch 1.8%
CAD 2,471.40 to 3,781.80
Production 11.3%
CAD 15,514.90 to 23,741.30
Holdback, 30 days after launch (10%)
CAD 13,730 to 21,010
GPU time for training and testing, at cost
CAD 300. Billed up front, at cost, outside the milestones

// questions

Questions we get about this.

Which document stores can the knowledge assistant read?

Shared drives and document systems with an interface. We confirm yours before we quote.

Does the knowledge assistant learn from our documents?

It searches them; it is not trained on them. Your documents stay in your account.

What if the knowledge assistant finds nothing relevant?

It says so, rather than inventing an answer.

// next

Not quite your project?

Tell us what you have in mind. We will come back to you with a range and the questions that would narrow it. Or book a call and talk it through.

AI knowledge assistant over a firm's own documents, on a private model | Atheron Network Labs