back_office_ops · finance · workflow

Wealthsimple's LLM journey: building a secure internal generative AI platform for employee productivity

Wealthsimple needed a secure way for employees to leverage LLMs without inadvertently sharing sensitive financial data with external providers, while also overcoming low adoption of an early internal gateway that employees perceived as an inferior copy of ChatGPT.

How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Gateway receives and proxies requests
Employees access the LLM gateway, available behind a VPN and gated by Okta, which proxies conversations to external LLM providers and tracks usage.
Tools used
LLM gatewayOpenAICohereOktaSlackNER modelllama.cppLlama 2MistralWhisperElasticsearchAirflowLangChainPythonStreamlitGeminiChatGPT
Outcome

Over 2200 messages are sent daily, close to a third of the company are weekly active users, slightly over half are monthly active users, and about 80% of all LLM usage flows through the internal gateway; almost everyone surveyed reported that LLMs significantly improved their productivity.

What failed first

Early nudge mechanisms that sent Slack reminders to employees using ChatGPT directly failed to change behavior, and the PII redaction model introduced accuracy and relevancy problems that degraded the user experience.

Results
Time savedclose to a third of the entire company
Volumeover 2200
Running since2023
Source

https://www.infoq.com/presentations/genai-productivity

How we source this →

Grounding & classification
Source type: technical build writeup
51 fields verified against source quotes, 1 dropped as unverifiable.
code generationcontent generationconversational aidocument aiknowledge searchragknowledge basesupport ticketbuilder submittedfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedfinancial servicesautomation rateemployee productivitythroughput increasetechnical build writeupback office opscustomer supportticket triageintake to triagerag answering