Spot the red flags
before diligence starts.
Upload a filing, a contract or a batch of documents and get a risk report in minutes. It’s scored across 15 M&A risk categories, and every finding is checked against the source text, so your team knows what to dig into first.
Uncompromising,
enterprise-grade security.
In M&A, confidentiality is everything.
Our architecture puts your data privacy above all else.
Defense in depth, down to the row
Client isolation isn't just app-layer logic — Postgres row-level security enforces it a second, independent time, so a bug in our code can't leak one client's data into another's.
Automatic PII stripping
SSNs, card numbers, emails, and phone numbers are redacted before any text reaches the AI model. Entity names stay intact for sanctions and adverse-media screening.
Confidential LLM partnership
Processed by Anthropic's Claude over encrypted TLS. Never used for training, retained only briefly for abuse monitoring — and already encrypted at rest on our side.
Encrypted at rest
Every byte of extracted text and AI analysis is locked down with Fernet symmetric encryption and HKDF key derivation — each client gets a uniquely derived key.
Clean rooms, enforced in code
The Clean Rooms add-on keeps competitively sensitive documents in a separate database schema with its own key. The deal team sees only what a lead or counsel releases, and every change, release decision and download is written to a hash-chained audit log. For gun-jumping and antitrust ethical-wall scenarios. How M&A clean rooms work →
Full audit trail
Every upload, analysis, report view, download and change is written to an append-only audit trail — who, what, and when — so you can show who did what with a document.
From upload to insight
in minutes.
Our intelligent document processor gets to work the moment you hit upload.
Upload
Drop in a PDF, Word document, CSV, or plain text file — one at a time, or 25 at once. From the portal, from your own systems over the REST API, or straight from your AI agent over MCP.
Analyze
The platform automatically identifies and prioritizes the highest-signal sections: material weaknesses, audit findings, risk disclosures, and litigation.
Review
Every finding is checked back against the source document and ranked into a verification plan — mark each one reviewed, flagged or dismissed. Export as PDF, DOCX, Markdown or JSON, or a batch as one watermarked ZIP.
15 M&A risk categories.
One composite score.
A comprehensive view of risk organized the way deal teams think — with blocking categories that halt a deal on detection. Each document is scored against one industry ruleset: M&A (15 categories), banking (8), pharma (7) or aviation (6).
See how each M&A category is screened →
M&A
⚠ Blocking — a score ≥70 in any of these categories flags the deal for mandatory review before proceeding.
Smarter analysis.
Real savings.
Manual document screening is slow, expensive, and inconsistent. A single engagement can cost thousands in analyst hours — and still miss critical red flags buried deep in a 200-page filing.
Speed
Complex documents analyzed in minutes, not days, with automatic prioritization of high-risk clauses and financial red flags.
Security
Per-client encrypted storage ensures zero cross-tenant exposure, with extracted text and findings encrypted at rest under keys unique to your engagement.
Accuracy
Measured, not asserted: 94.1% precision on the deal-blocking gate across 20 annotated SEC filings anyone can re-fetch. Sample size, confidence intervals and the findings it missed are published in full. Accuracy & benchmarks →
Value
A fraction of the cost of a traditional diligence engagement, with transparent subscription pricing and no hourly billing surprises.
Frequently asked questions
What is Anweshna?
Anweshna is an AI-powered M&A due diligence platform that scores uploaded deal documents (PDF, DOCX, TXT, CSV) across 15 M&A risk categories — financial, legal, HR/labour, sanctions & AML, regulatory/merger control, IP & licensing, and more. Banking, pharma and aviation deals are scored against a dedicated industry ruleset instead (8, 7 and 6 categories respectively). It flags a deal for mandatory human review when any blocking category scores 70 or above.
How is Anweshna different from a manual due diligence review?
Manual document screening for a deal can take analyst teams days to weeks and still miss risks buried in long filings. Anweshna processes the same documents in minutes, automatically prioritizing high-signal sections — material weaknesses, audit findings, risk disclosures, litigation. A filing too large to read in a single pass is not trimmed to fit: it is split into chunks and every chunk is scored.
Does Anweshna replace a lawyer or independent due diligence?
No. Anweshna scores documents for risk signals — it does not replace legal review, independent verification, or expert due diligence. It's designed to accelerate and structure the initial screening pass that a deal team would otherwise do manually, surfacing what to investigate further.
What document types does Anweshna support?
PDF, DOCX, TXT, and CSV files can be uploaded one at a time or in a batch of up to 25 files. Reports are downloadable as Markdown, JSON, PDF, or DOCX depending on your plan. A whole batch comes down as a single ZIP, and every report carries a provenance watermark — who downloaded it, their role, and the timestamp in their own timezone — so a file filed into a data room identifies itself.
How is client data secured?
Every client's extracted text and analysis findings are encrypted at rest with a per-client key. Documents are processed by Anthropic's Claude over encrypted TLS and are never used to train AI models.
Can Anweshna be used as an M&A clean room?
Yes, for the access-control part of one, with the Clean Rooms add-on. A clean room holds its material in a separate database schema with its own access role and encryption key, and gives each member a room role — lead, counsel, clean team or deal team. The deal team never sees the documents, and nothing reaches the deal team until a lead or counsel releases it through the Release Airlock. Every change, release decision and download goes into a hash-chained audit log. This is the environment a clean team works inside, for gun-jumping and antitrust ethical-wall scenarios. Anweshna does not provide clean-team services: we supply the room, the risk screening and the audit trail, not neutral third-party personnel. The add-on is available on the Growth and Pro plans and custom on Enterprise.
Can I use Anweshna from my own systems or an AI agent?
Yes, on every paid plan. Besides the web portal there is a REST API for server-to-server automation, and a hosted MCP server at https://anweshna.com/mcp/ that exposes screening as tools inside Claude, ChatGPT and other agentic deal workflows. An agent can screen a single document, fetch an analysis it has already run, submit a batch of up to 25, and poll that batch's status. The agent acts as your account, not around it — same API key, same plan limits, and the same role and clean-team restrictions a person would hit. API & MCP integration guide →
Data isolated and encrypted.
By design.
Security isn't an afterthought — it's baked into the architecture from the ground up. Every client's documents and findings are encrypted at rest under a key derived uniquely for your engagement.
No credit card required