Process-mining agents found 20 steps and 7 loops in a workflow documented as 7 steps
vasuman · x · 2026-10-02
A Forward Deployed Engineer shares how they audited a real workflow at a $5B public software company with 150+ products:
- Read the process doc: theirs listed 7 steps (build quote → submit → Deal Desk → approve → send → negotiate → sign).
- Interview more than one person—a single informant can give you the wrong story.
- Run process-mining agents on the CRM: they found 20 steps and 7 loops—61% of requests get kicked from step 4/5 back to step 1, Legal sends requests back 12% of the time, and 30% of new quotes require re-approval.
- Map the real workflow: the CRO and CFOs said, "you understand our department better than we do."
Key advice: do this audit before applying AI, so you know which process is actually broken.
More from coding & agent
- Reddit: vLLM and llama.cpp already ship training-free zero-shot classifiers via grammar-constrained output — Altruistic_Heat_9531 · 2026-10-02
- Dev Builds a Boxing Trainer App in One Shot with "5.5" — ZeroStateReflex · 2026-10-02
- Dev Wires Grok Bot and OpenAI Dot Into Voice Interfaces for a Hermes Agent Assistant — alexcovo_eth · 2026-10-02
- Translating an entire book with DeepSeek: pennies and under an hour, decent quality — teortaxesTex · 2026-10-02
- OmniSeek turns Omni-LLMs into agents that actively seek audio-visual evidence — Haibo Wang · 2026-10-02
- Microsoft's ActiveSaddler Uses Automated Curriculum Learning to Boost Agent Harnesses by 7.5 Points — microsoft · 2026-10-02