Case study · GTM / RevOps

Outbound-to-CRM pipeline

A GTM system that turns cold-email events into CRM contacts, deals and team alerts, and keeps opt-outs, lead rotation and follow-up enrolment in sync without manual work.

Built for KodifyClient engagementGTM / RevOps
Engineering facts
10 automated jobs from one container image
5-layer reply classifier before anything becomes a deal
60 automated tests
  1. 1The problem

    Replies, opt-outs and interested leads from outbound campaigns had to be copied into the CRM by hand, so deals were missed, opted-out contacts risked being emailed again and nobody trusted the pipeline view.

  2. 2Our contribution

    Designed and built by Kumar Sahil.

  3. 3What we built

    One container image with 10 run modes on serverless jobs: a webhook receiver for campaign events, opt-out sync, lead rotation, follow-up enrolment, a reconciler, monitoring and credit alerts. A 5-layer reply classifier (platform flag → keywords → LLM → statistical model → keyword safety net) decides which replies become deals, and secrets live in a managed secret store.

  4. 4The result

    In production. Results shared on request.

Technology
Python
Flask
Docker
Serverless jobs
Instantly
HubSpot
Slack
LLM classification
More work

Other systems we’ve built

Open source

Reply Classifier Cascade

A cost-ordered classifier for English and Dutch cold-email replies that decides which replies become CRM deals, built so a false “hot deal” needs several layers to agree.Read the case study
Open source

LLM Gateway

A provider-neutral LLM gateway: code asks for a job, the gateway routes it through a chain of models with safe failover, a deadline per call, and the cost of every call written to a ledger.Read the case study
Open source

LLM Eval Gauntlet

Invariant-checked evaluation suites for tool-calling agents that run in CI and fail the build when an agent invents a number, leaks its prompt or obeys an injection.Read the case study
Next step

Find the workflow worth automating first.

Tell us about one process that costs your team hours every week. We will tell you honestly whether AI or automation is the right fix, and what a first build would take.

Not ready for an assessment?

Ask one question about a task you’d like automated. You’ll get a straight answer by email.
0/2,000

Answered personally by Kumar Sahil. No newsletter, no sales sequence. Privacy.