Sales Automation
Modjo's Atlas-backed conversational-intelligence platform was hitting 100% CPU saturation because inefficient connection handling, unclosed sessions, and large requests were driving performance bottlenecks and unnecessary infrastructure cost.
Reported impact
After minor code changes, Modjo cut CPU usage from 100% to 20%, reduced query targeting from 40,000 to 1, downsized its Atlas cluster from M60 to M30, and lowered monthly Atlas costs by 75%.
Modjo uses MongoDB Atlas as the operational database behind its conversational-intelligence platform, which ingests customer-call data from Zoom, Skype, and Google Meet and runs separate transcription and speaker-identification pipelines. Atlas stores orchestration state, logging, and next-step decisions; writes reach Atlas through a custom API built on AWS Lambda and API Gateway while downstream services read directly. As load grew, inefficient connection management, unclosed sessions, and large per-request datasets drove 100% CPU saturation on an oversized cluster. MongoDB Professional Services first reworked schema and indexing, then traced persistent hot spots to per-operation client creation and leaked sessions, guiding code changes that reused a single client and explicitly closed sessions after retrieval.
Conversation data from Zoom, Skype, and Google Meet feeds Modjo's transcription and speaker-identification pipelines. Orchestrator services write task states and next-step decisions through AWS Lambda and API Gateway to MongoDB Atlas, while downstream services read Atlas directly to drive analytics and the end-user application.
Modjo’s solution is designed to process calls using two data science pipelines; one for transcription and one for speaker identification.
Within a few hours, CPU usage dropped from 100% to 20%
Writes to the database are performed via a custom API (Lambda + API Gateway), while reads are direct.
resulted in a 75% reduction in monthly MongoDB Atlas costs
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
The legal notice lists "Head office address: 59 Avenue Sainte-Foy, 92200 Neuilly sur Seine," placing the headquarters in France.
Related
MongoDB Sample Sep 30, 2026 MongoDB MongoDB needed to stop sales reps from spending 4 to 6 hours per account or deal cycle manually stitching together customer context from Salesforce, email, meeting transcripts, and documents before they could plan… Holly currently runs on 30 agents. Atlas Agent Engine MongoDB Sample Aug 11, 2026 Scalestack Scalestack needed to unify first-party, third-party, and zero-party go-to-market data so sales teams could stop manually researching prospects and act on AI-generated recommendations faster. Customers saw a 40% increase in rep productivity. MongoDB Atlas MongoDB Sample May 18, 2026 LekoTech LekoTech needed to digitize millions of unique car parts to promote reuse but faced latency of up to 35 seconds with its existing search system. LekoTech needed to digitize millions of unique car parts to promote reuse but faced latency of up to 35 seconds… MongoDB Atlas · MongoDB Atlas Search · MongoDB Atlas Vector Search 