Revolution AgencyRevolution Agency

Delivered by our team through its European experience (Evolution Agency, France).

Case study: IT interim management, France

From client brief to candidate file, every conversation ends with a document.

For Infortive, an IT interim management firm, our team deployed four agents on its two Bubble platforms: two for taking briefs, on the client side and the internal side, and two for writing candidate files.

4

agents in production: two for briefs, two for candidate files

2

Bubble platforms equipped: the client portal and the internal portal

2

documents written for every validated brief: the engagement letter and the internal summary

Methodology: the results on this page are qualitative. They come from operating the tools and from the testing done before each launch. We publish client-specific percentages only once they are measured and approved.

Context

Infortive is an IT interim management firm. It presents interim managers and freelancers to its clients, and runs on two Bubble platforms: one portal for its clients, one for its internal team.

Two steps weighed on the team: collecting a complete mission brief, and writing each candidate's file. The goal was to automate without losing quality. The starting point:

  • Thousands of CVs in their Bubble platform
  • Brief taking that required senior IT sales experience
  • Candidate files that took several days to write
  • Two file formats to follow: freelance and interim management

Challenge

Automate two steps that required experience, without losing quality, inside the tools the team already used.

1

A complete brief

Collect the context, scope, target profile, schedule constraints, budget and deliverables, whether the brief comes from the client or from an account director.

2

Files true to the documents

Write the file from the CV, the AssessFirst report and the engagement letter, in the firm's format: freelance or interim management.

3

Fit the existing setup

Work inside the two Bubble platforms, save the documents to Google Drive and keep the internal portal's conversation log up to date.

Solution: two pairs of agents and a chain of automations

Four conversational agents built and maintained by our team, connected to n8n automations that read the documents, fill in the templates and file the result in Google Drive.

  1. 1

    The brief, client side

    On the client portal, the client describes the need to the agent. They can attach documents, read by OCR (Mistral) and added to the conversation's context. The agent rephrases and offers summaries along the way. On validation, a Google Drive folder is created for the session.

  2. 2

    The brief, internal side

    On the internal portal, the account director picks up the opportunity already recorded in Bubble. The agent reads it, completes the brief with them, then updates the opportunity.

  3. 3

    Engagement letter and summary, written automatically

    At the end of every validated brief, the automation copies two Google Docs templates into the mission folder, the engagement letter and the internal summary, and fills them in. The matching conversation on the internal portal receives the title, the folder link and the conversation reference.

  4. 4

    Candidate files, in two formats

    On the internal portal, the recruiter uploads the CV, the AssessFirst report and the engagement letter. Mistral OCR reads the documents, then the agent asks for what is missing: availability, day rate, mobility, strengths, Infortive's recommendation.

  5. 5

    A file written in the firm's template

    On validation, two AI agents (Claude and GPT) structure the profile and the work history. The file is filled into the Google Docs template for the chosen format, freelance or interim management, experience table included, then saved to the delivery folder.

  6. 6

    Models chosen per agent, and a voice

    Each agent uses its own models, Claude (Anthropic) and GPT (OpenAI), depending on its task. Replies are read aloud with a synthetic voice (ElevenLabs). An error on a critical step triggers an email alert to our team.

  7. 7

    Batch CV processing

    Alongside the four agents, we built a batch process with five AI agents: mission types, roles, contact details, sectors and work history. It was used to extract and structure hundreds of CVs into a reference spreadsheet.

Results

No percentages on this page: we only publish a client rate once it is measured and approved. The results below are observed in operation.

Four agents in production on both portals, client and internal

Every validated brief produces its engagement letter and internal summary, in a Google Drive folder dedicated to the mission

Candidate files written in the firm's template, freelance or interim management, ready to share with the client

At the end of every brief, the internal portal's conversation log receives the folder link and the conversation reference

The honest framing: the agents do not pick candidates. They collect the brief, read the documents and write the files. The recommendation stays with the Infortive team, and the agent asks for it before writing the file.

Stack

Conversational agents embedded in BubbleBubble Data APIn8n automationsMistral OCRClaude (Anthropic) and GPT (OpenAI) modelsElevenLabs synthetic voiceGoogle Drive and Google DocsGoogle Sheets

Your briefs and files follow a template. Your agents can write them.

Evaluate your project free with our interactive tools, or book a call: we look at your briefs, your files and your current tools, and tell you what an agent can handle, and what it should leave to your team.

Book a call
NovaVoice guide
Ready