Case study · AI engineering
Three funds, roughly 200 documents, and zero numbers made up.
India Avenue is the only investment firm in Australia and New Zealand devoted entirely to India, run out of Sydney and Mumbai since 2015. We built the assistant that answers their investors at any hour, using only the figures inside the fund documents, with a link back to the source every time.
What it changes: Reading one published number used to take four steps, and a year of performance meant four factsheets compared by hand. Now it is one question, and every question asked is one the firm can finally see.
- Client
- India Avenue Investment Management
- Sector
- Funds management, Sydney and Mumbai
- Status
- Live in production, as the Avenue Virtual Assistant
- What we built
- An investor support assistant
- Funds covered
- Equity Fund, India 2030 Fund, Active ETF
- Corpus
- Roughly 200 documents across the three funds
- Grounded retrieval
- Purpose built tools
- Chart from factsheet
- Source citations
- No advice, no forecasts
- 50 question eval suite
- Model agnostic
3
Funds it answers for
~200
Documents behind them
50
Questions it had to pass
100%
Of answers from approved documents
The ask
India Avenue holds an Australian financial services licence. Its investors are advisers, family offices and people with their own savings in the funds. A wrong return, or a stray forecast in an inbox, can mislead an investor and put the licence at risk. So the bar was simple. The assistant could be helpful, but it could never be loose with a number or an opinion.
They asked for two things. Sound like India Avenue, not like a chatbot. And be right, and say so plainly when it is not. Underneath both sat something quieter: they had no way of knowing what visitors were actually looking for on their own site.
The hard part: the figures were pictures
What went wrong
Much of the data does not exist as text. It is locked inside images of charts and tables. Ordinary retrieval could not read it, so the agent answered from whatever words happened to sit nearby. Accuracy was poor, and the prompt was not the reason.
What fixed it
Not better prompting. We reworked the retrieval and split roughly 200 documents into separate, purpose built tools. Each handles the document type it suits, so the agent pulls a figure from a source that actually contains it.
Every answer had to be
Sourced
Every figure comes from an approved fund document, with a link back to it.
From documents, not memory
Pulled fresh from the factsheets and research, never recalled by the model.
No advice, no forecasts
It shares data and explains strategy. It never recommends and never predicts.
How we worked
- Discovery
- Learned the funds, the share classes and the documents investors actually ask about.
- Proof of concept
- One assistant, trained on their own factsheets and research notes. Nothing from outside the documents.
- They marked it
- India Avenue sent 50 real visitor questions, then marked every answer right or wrong with the reason. We tuned and ran it again.
- Production
- Live on the site as the Avenue Virtual Assistant, answering fund, performance and research questions.
What it does now, and what we left out
Asked for advice it will not give
“I can certainly provide the detailed historical performance data and strategy documents for our funds. For personalised guidance I recommend speaking with your own financial adviser, and our support team can provide any fund documents they may require.”
What we left out on purpose
The brief had the agent emailing documents on request. The portal that holds them opens no API to outside software, so a half connected version would have behaved like a dead end. It stayed out of version one. The agent takes the request and hands it to the support team.
“AvestaLabs worked closely with us to identify a high-impact AI opportunity that aligned perfectly with our business. The AI agent they delivered has added real value to our business. Their collaborative approach, technical expertise and strong command of agentic AI made the entire engagement seamless and impactful.”
The job transfers
Anyone answering on regulated numbers
Wealth advisers
Super funds
Private banks
Anyone with data trapped in documents
Insurers
Fund administrators
Any firm whose figures live in PDFs
More ai engineering work
AI engineering · Life insurance · Australia
They stopped choosing which calls to hear
Life Insurance Direct sells life cover on the phone. We built the system that reads every one of those calls, scores both sides, and hands the team a review queue in risk order instead of a sample.
- 9,218 in four months, both sides
- Calls read
- Live, in their own cloud
- Status
Next
Tell us which part of this looks like your problem.
We will walk you through how it was built, who built it, and what it took to keep it running afterwards.