Estimator

Which question are you asking?

Pick the one closest to your situation. You get the same published band you would find on the pricing page, filtered to the arc that applies — and the assumptions that decide whether it holds.

Start here

Intelligence — published bands

Agents, retrieval, voice and document AI over tenant-scoped data, with evaluations in CI, cost ceilings per run, and a human step before anything commits the business.

Published price bands and the assumptions behind them
BandFromWhat it buysTypical durationAssumes
Retrieval Pilot₹6,00,000One corpus indexed with tenant scoping, a 50-question golden set, and accuracy measured before and after.3 weeks
  • Up to 5,000 documents in text-extractable formats
  • One tenant boundary; cross-org sharing rules are separate scope
  • Accuracy measured against a question set your team signs off
Agent Build₹15,00,000One production agent: tools, guardrails, escalation, evaluations in CI, and a cost cap per run.8 weeks
  • The process the agent takes over is documented, or observed during Envision
  • Model and inference spend is billed at cost, outside the fee
  • Two systems integrated; each additional system is quoted separately
Agent Layer₹34,00,000Several agents on shared retrieval with model routing, approval queues and per-agent cost reporting.14–18 weeks
  • Three to five agents defined by a signed build plan
  • Source data sits in systems exposing an API or a database we can read
  • Voice channels require a telecom account in your name
Take this to scopingWhat this involves