API and search

01CHANNELS · API AND SEARCH

The platform over an API

Parsewise is an AI assistant that answers on your company data. The API opens the same platform: answers, search and actions — inside your own apps, bots and services.

Website
your frontend
Mobile app
iOS · Android
Telegram bot
messengers
Voice
telephony
Search
endpoint
Streaming
SSE
Webhooks
events
CRM
integration
Your case
open slot
Parsewise
02TWO ENDPOINTS

Two methods — dialogue and search

The API surface is deliberately small. Behind two calls stands the whole platform: knowledge base, skills, connectors and analytics.

Get a key
POST /chat

Dialogue

A question and an answer from your data, with dialogue memory, skills and handover to an agent.

POST /search

Search

Semantic search across the knowledge base: it finds by meaning, not by matching words.

SSE

Streaming

First token under a second — the answer types itself into your interface in real time.

03IN THE BOX

More than a model

There is a platform behind the API. Data is uploaded in the admin without code, and the API returns finished answers — with sources, actions and live data.

PLATFORM/01

Knowledge base

Upload documents, your site, products and profiles in the admin — Parsewise indexes them and answers from them.

PLATFORM/02

Skills

Booking, request, quote — ready actions are called from the dialogue and return the result into your interface.

PLATFORM/03

Connectors

MCP tools and data hooks: the assistant pulls live prices, stock and statuses from your systems.

PLATFORM/04

Admin

Keys, logs, dialogue history, ratings and analytics — all visible without a line of code.

DEMO · CONSOLE1:20

Key → POST /chat → answer

The first API request in a couple of minutes, straight from the console

04INTEGRATION

The integration took a day: a key, a /chat call — and the assistant was already answering inside our app.

Anton

Backend developer

05SLA · UPTIME

Reliability in numbers

An SLA is a measurable metric. We keep uptime and latency under control and put the numbers on dashboards you have access to.

01

API uptime
per SLA on the production contour

02

First token
streamed response

03

Search p95
semantic /search

04

Coverage
monitoring and alerting

Uptime by month · %
UptimeSLA target 99.9%
Latency · by method · ms
p95p50
06WHAT IS INCLUDED

What the API includes

One key covers the whole contour. From the models and the knowledge base to analytics and quotas — with transparent request statistics every month.

Requests over time · by type
ChatSearchStreamingWebhooksTotal
MODELS
Answers · embeddings · reranker
DATA
Documents · products · profiles
SKILLS
Forms · actions · use cases
CONNECTORS
MCP · CRM · 1C · warehouse
ANALYTICS
Logs · history · ratings
LIMITS
Keys · quotas · billing
07FAQ · FOR DEVELOPERS

Developer questions

/chat runs a dialogue: it answers from your data, keeps the context, calls skills and can hand over to an agent. /search is semantic search across the knowledge base: it returns relevant fragments by meaning, not by matching words.

ROLLOUT

Judge Parsewise on your own data

Free plan, no card required. Upload your material, take a key and judge the answer quality on real data.

/01

Sign up

Create an account in a minute — no card, free plan forever.

/02

Data

Upload documents, products and profiles — the knowledge base is ready.

/03

API key

Copy the key from the admin — /chat and /search are open.

/04

First call

Call /chat and judge the answer quality on your own material.

REQUEST

Tell us your task

PORTFOLIO BY TYPEBY YEAR

Projects by type grow year over year

MVPRedesignAISupportTotal

MVP, redesign, AI and support — cumulative

STRENGTHSPROFILE

The studio profile across key axes

Speed, quality, transparency, engineering

PROJECT PHASESOVER TIME

Research, design and build overlap

Parallel streams — not a waterfall