How Cando Rail Cut Risk-Assessment Time 83% With AI

A rail safety team spent 4–16 hours of expert meetings on every risk assessment. An AI now drafts and documents them live — freeing 20+ hours and cutting scale-up time 83%.

83%

Reduced time to scale program 6x

Implementation Time

Chris Taylor

CEO & Co-Founder

Fractional AI

Connect ↓

The Challenge

Cando wanted to scale its risk assessment program sixfold but faced heavy administrative burden. The existing process required 4-16 hours of meetings per assessment for up to 10 subject-matter experts, plus 2+ weeks of back-and-forth coordination that pulled experts away from safety-critical work.

What They Built

Fractional AI built Peter the Safety Agent, an AI assistant with two workflows: automated draft generation for pre-meeting preparation, and a real-time in-meeting voice assistant that documents and updates assessments live. It draws on historical risk assessments, safety procedures, incident reports, and training data.

The team used GPT-4.1 for drafting and the OpenAI Realtime API for voice, orchestrated through LangChain agent flows with RAG, integrated with MS Teams and Recall.ai for meeting transcription. A key insight was that human-written assessment titles outperformed similarity-based RAG retrieval for selecting relevant context.

AI Role

Peter operates in two modes. Before a meeting, it drafts a full risk assessment from historical data for experts to refine. During the meeting, a real-time voice assistant listens, documents decisions, and updates the assessment live, letting subject-matter experts focus on judgment rather than note-taking.

AI Model

GPT-4.1 (drafting), OpenAI Realtime API (voice), LangChain agent flows with RAG, MS Teams, Recall.ai for transcription

Infrastructure

Microsoft Teams (meeting environment) • Recall.ai (meeting transcription) • Knowledge base of historical assessments, safety procedures, incident reports and training records

Integration Points

Live Microsoft Teams meeting integration • Recall.ai transcription piped into the agent • RAG retrieval over the assessment knowledge base • Draft-generation workflow feeding pre-meeting prep

Impact

20+ hours saved per assessment

Each risk assessment saves over 20 hours of subject-matter-expert time versus the prior manual process.

83% faster to scale 6x

Reduced the time needed to grow the risk assessment program sixfold by 83%.

Under $0.05 per draft

Automated draft assessments are generated at less than $0.05 per generation.

Technology Utilized

Implementation Complexity

Time Savings

Risk & Compliance

Best Fit For

Best fit for safety- or compliance-heavy operators who run repetitive, expert-intensive assessment meetings and need to scale them without adding headcount.

Chris Taylor

CEO & Co-Founder

Fractional AI

CEO & Co-Founder of Fractional AI, helping PE firms and portfolio companies implement AI workflow automations, product features, and diligence at scale.

Get an intro

Industry:

Transportation & Logistics

Business Function:

Operations

Company Size:

Project Cost:

Cost not disclosed

Ownership:

Organization Type:

AI Pattern:

Conversational AI (Chatbot / Agent)

Knowledge Management & Search (RAG)

Value Type:

Time Savings

Risk & Compliance

AI Model:

Frequently Asked Questions

How did Cando Rail cut risk-assessment time by 83%?

Through an AI safety agent called Peter that drafts risk assessments before meetings and documents them live during them. Grounded in Cando Rail's historical assessments, safety procedures, and incident reports, it removed the manual bottleneck and cut the time to scale the program sixfold by 83%.

What AI tools and models did Cando Rail use for its safety agent?

The agent used GPT-4.1 for drafting and the OpenAI Realtime API for voice, orchestrated through LangChain agent flows with retrieval-augmented generation (RAG). It integrated with Microsoft Teams and used Recall.ai for meeting transcription.

What results did Cando Rail achieve with its AI safety agent?

Three main outcomes: over 20 hours of subject-matter-expert time saved per assessment, an 83% reduction in the time needed to scale the program sixfold, and automated draft assessments generated for under $0.05 each.

How long did it take to see results?

A specific implementation timeline was not disclosed. Impact was measured per assessment once the agent was live, with each one saving over 20 hours of expert time.

Who is this AI safety agent approach best for?

It is best suited to safety- or compliance-heavy operators that run repetitive, expert-intensive assessment meetings and need to scale them without adding headcount.