§ 01 · Launching in 48 hours · Chennai, India

skryb

by DemanualAI · Chennai, India

Your company's entire unstructured brain — queryable in plain English.

Every meeting note, policy, vendor contract, and voice recording your team has ever produced — findable in seconds, with a source citation attached to every single answer.

Launching in
--HRS
--MIN
--SEC
Request pilot accessSee what it solves ↓

25–500 employee companies · India INR pricing · UAE USD pricing · Starter from ₹15,000 / month

§ 02 · The problem

The documents exist. They're just unfindable.

Every mid-market company runs on documents and conversations no one can find when they need them. Meeting minutes get saved as PDFs and forgotten. Policies live in Word files on someone's laptop.

"I know we decided this six months ago but I can't find which meeting."

CEO · 80-person logistics company

"What's our policy on travel reimbursements?"

HR Head · 150-person fintech

"What exactly did we agree to in the vendor contract with [X]?"

COO · 60-person manufacturing firm

"Did we ever discuss this project in a board meeting?"

CFO · 200-person retailer

"Send me everything we have on that client from the last two years."

Sales Lead · 45-person SaaS company

§ 03 · What skryb does

One place to ask any question about anything your company has ever written or recorded.

Get an answer with the source attached — document name, page number, or audio timestamp — every single time.

01

Upload anything

PDFs (text + scanned), Word, PowerPoint, spreadsheets, plain text, audio. Drag-and-drop or bulk import. Up to 100 MB per file.

02

Automatic deep indexing

Every document is extracted, chunked into three levels (2,048 / 512 / 128 tokens), embedded with BGE-M3, and enriched with summaries, keywords, and named entities. A 10-page PDF is ready in under 90 seconds.

03

Ask in plain English

Type any question on web or WhatsApp. Hybrid retrieval runs across your entire workspace — dense vector search plus BM25 full-text, fused with Reciprocal Rank Fusion.

04

Cited answer, every time

Every answer links to the exact document, page number, or audio timestamp. If confidence is below threshold, skryb says so. Hallucination is not a product tradeoff we accept.

Citation SLA

Every answer includes the source document name, page number or audio timestamp, and a confidence indicator. If retrieved context doesn't contain the answer, skryb returns "I don't have enough information to answer this confidently" . Never a fabricated answer. This is non-negotiable.

§ 04 · Supported formats

Every format your team already uses.

PDF
Text-native + scanned (OCR)
DOCX / DOC
Microsoft Word
PPTX / PPT
Slides + speaker notes
XLSX / CSV
Spreadsheets as documents
TXT / MD
Plain text and Markdown
MP3 / WAV / M4A / OGG
Audio — transcribed with diarization
EML / MSG
Email threads
v1.1
PNG / JPG
Images with text (OCR)
v1.1

Max 100 MB per file · Max 4 hours per audio file · Google Drive and SharePoint connectors in v1.1

§ 05 · How it's built

Production-grade RAG architecture. Not a wrapper.

Every engineering decision is documented in the technical spec and grounded in benchmarks. Nothing is hand-wavy.

Embedding
BGE-M3 by BAAI

Self-hosted on AWS Lambda (Docker container). 1,024-dim vectors. 8,192-token context window. Covers English, Hindi, Tamil, and Arabic without a model change.

Retrieval
Hybrid dense + sparse → RRF

pgvector HNSW cosine similarity + PostgreSQL BM25 full-text search. Signals merged with Reciprocal Rank Fusion. Proper nouns, dates, and contract numbers score correctly.

Chunking
LlamaIndex hierarchical

Three-level hierarchy: 2,048 / 512 / 128 tokens. AutoMergingRetriever promotes grandchild chunks to parent context before synthesis. No information loss.

LLM routing
AWS Bedrock Nova Pro / Lite

Nova Pro for complex multi-document synthesis (~30% of queries). Nova Lite for simple factual recall (~70%). Smart-routed per query. Blended cost ~$0.002/query.

Grounding
Strict grounding prompt, always

Every model invocation is instructed to answer only from retrieved context, cite the source, and return a refusal if confidence is below threshold. No exceptions.

Storage
Supabase + pgvector + AWS S3

Vectors, metadata, conversations, and audit logs in Supabase Postgres. Raw files in S3. Row-Level Security enforces tenant isolation at the database level.

Metadata
5-extractor pipeline

Title · Summary (with prev/next context) · Questions answered · Keywords · Named entities (people, orgs, amounts, dates). Runs on every chunk level at ingest.

OCR
pdfplumber + AWS Textract

pdfplumber primary. Auto-fallback to Textract when quality check fails. Full table/form analysis is feature-flagged to Growth and Boardroom tiers.

Audio
Amazon Transcribe

Batch mode. Speaker diarization for multi-speaker recordings. Timestamps preserved per chunk so every cited answer links to the exact second.

§ 06 · Audit and compliance

Every action logged. Every answer traceable.

The audit trail is append-only and retained for 12 months minimum — exportable as CSV on Growth and Boardroom tiers. Built for regulated buyers: BFSI, listed companies, company secretaries managing board compliance under SS-1 and SS-2.

·Every document upload — who, what, when
·Every question asked and every answer returned
·Every permission change and member invitation
·Every document deletion
·Append-only log — no edits, no deletions
·12-month minimum retention, CSV export on Growth+
·AES-256 at rest · TLS 1.2+ in transit
·DPDP-ready (India) · PDPL-ready (UAE)
·Data stays in-region — contractual commitment
·SOC 2 Type I target: month 9 post-launch
§ 07 · Pricing

Priced for Indian and UAE companies. Annual contracts, monthly billing.

Starter
Talk to us →
·1 workspace
·Up to 25 users
·500 documents
·10,000 queries / month
·English language
·Q&A on Web + WhatsApp
·Textract OCR fallback
Growth
Talk to us →
·1 workspace
·Up to 100 users
·5,000 documents
·50,000 queries / month
·English + 1 regional language
·Full OCR table extraction
·Audit log CSV export
·Per-document access control
·Priority support
Boardroom
Talk to us →
·Single-tenant deployment
·Up to 500 users
·50,000 documents
·500,000 queries / month
·All languages
·SS-1 / SS-2 compliance pack
·WhatsApp document upload
·Dedicated CSM
·Custom DPA

All tiers · Annual contracts · Monthly billing · 2 months free on annual prepay

§ 08 · Built by DemanualAI
DemanualAI
Chennai, India

An AI products company building for the companies that are actually building India and the UAE.

We're engineers first. Services revenue funds product development — which means we know what companies at this scale actually need. skryb is Module 1 of 5 in the DemanualAI Org Intelligence OS.

We don't call ourselves revolutionary. We write software that works, price it for your market, and put our names on it.

Talk to the founders →
FoundedChennai, India
Productsskryb (Module 1) · Dataviz · LegalPro · DevHire
Target marketsIndia (INR-priced) · UAE (USD-priced)
Company size fit25 – 500 employees
Data residencyAWS Mumbai (ap-south-1) · AWS Bahrain (me-south-1)
Compliance trackDPDP India · PDPL UAE · SOC 2 Type I (month 9)
Uptime target99.5% in v1 · 99.9% committed in v2
Latency target (P95)Question → answer ≤ 6 seconds
v1 ship timeline8 – 12 weeks from kickoff
§ 09 · Pilot cohort open

Ready to stop searching and start finding?

We're onboarding a small pilot cohort of design partners before public launch. If you run a 25–500 person company in India or UAE and you live in documents, email us directly.

founders@demanualai.com

Pilot pricing: ₹15,000 – ₹40,000 / month · Annual contract · Ships in 8 – 12 weeks