OpsGenie AI — Enterprise Operations Hub

The Challenge

A mid-size professional services group operating three firms — a 160-person management consultancy, an 80-person legal advisory, and a 270-person staffing agency — was hemorrhaging productivity to administrative overhead. A time-audit conducted before the engagement revealed that team leads averaged 3.1 hours per day on non-billable administrative tasks: writing meeting recaps, manually delegating action items, processing HR form submissions, chasing approvals, and context-switching between Slack, email, Google Calendar, and their project management tools. Across 500 employees at an average fully-loaded cost of $85/hour, this represented $13M+ in annual unrecoverable administrative drag. The group had tried conventional project management tools and automation platforms but found them too rigid for the dynamic, cross-firm coordination their operations required. They needed AI that could understand context, not just execute predefined workflows.

Our Custom Solution

OpsGenie AI is a multi-agent operations platform with four specialized AI agents coordinated by a LangChain orchestrator. The Meeting Agent uses OpenAI Whisper to transcribe all Zoom/Google Meet calls in real-time, then passes transcripts to Claude for structured summarization: decisions made, action items with owners, blockers, and next meeting agenda. Summaries are auto-posted to the relevant Slack channel and Notion project page within 4 minutes of call end. The Delegation Agent monitors action items, tracks completion against deadlines, and sends context-aware reminders ("The Morgan proposal deck is due tomorrow — you mentioned in Monday's call that Sarah was handling slides 12–18") rather than generic nudges. The HR Agent automates the full lifecycle of routine HR requests: PTO approvals routed by policy rules, expense report pre-validation, onboarding document collection, and benefits enrollment reminders — all handled conversationally via Slack DM. The Command Dashboard (Next.js + Supabase) provides executive visibility: cross-firm task velocity, admin time saved per team, upcoming deadline risk matrix, and headcount utilization heatmaps.

System Architecture & Technologies

• LangChain multi-agent orchestrator coordinating 4 specialized agents with shared memory context
• OpenAI Whisper real-time meeting transcription with speaker diarization via pyannote.audio
• Claude API for structured summarization, action item extraction, and HR request interpretation
• Slack API for conversational agent interface and automated channel posting
• Google Calendar + Zoom webhooks for automatic meeting session detection and recording
• Notion API for structured meeting note and project documentation sync
• Supabase (PostgreSQL + realtime) for cross-firm task tracking and agent state persistence
• Next.js command dashboard with Supabase realtime subscriptions for live operations data

Technologies: Claude API • OpenAI Whisper • LangChain • Next.js • Supabase • Slack API • Google Calendar API • Notion API

Measurable Results & Outcomes

• 60%: Reduction in per-employee administrative time (3.1h → 1.2h/day)
• 500+: Employees across 3 firms actively served by the platform
• 4 min: Median time from meeting end to structured summary in Slack
• $7.8M: Annualized productivity value recovered at blended $85/hr cost rate
• 94%: Action item capture accuracy versus manual meeting notes
• 3: Enterprise firms fully deployed across all 4 agent modules

Client Testimonial

"We got back 60% of the time our senior team was spending on admin work. The meeting summaries alone changed how we operate — every action item is tracked, nothing falls through the cracks."

Explore More Case Studies

• SmartSite Vision — Construction Site Safety AI (73% incident reduction)
• MediFlow — Clinical Document Intelligence (4h+ saved per physician/day)
• BuildBot — Construction Project Manager AI (31% on-time improvement)
• CareIQ — Patient Triage & Readmission Predictor (41% readmission reduction)