Cross-FunctionalIT & Security
Document intake. Filed right, the first time.
Classifies and tags every incoming document against your enterprise taxonomy at the point of intake, with a full audit trail — so every downstream system finds it filed correctly the first time. Built for teams ingesting thousands of documents a month across formats.
13 cited fields per document. Built on Sphere AI Foundry.
doc_classification — file #DC-10552LIVE
00:00Ingest → inbound file, unclassified.pdfread
00:02Type detection → matches invoice template, 96% confidencepassed
00:03Taxonomy match → Finance > AP > Vendor Invoicepassed
00:04Metadata extraction → vendor name ambiguous, 2 possible matchesflagged
00:05Disposition → Route to reviewer, 1 field to confirmrouted
classification time 4sfields cited 13/13reviewer sign-off required
Organizations around the world trust us
Built on Sphere AI Foundry·Self-learning taxonomy match·21 years, 300+ clients
Manual filing doesn't scale with document volume
A single incoming document means someone opening it, figuring out what it actually is, and filing it under the right taxonomy node so the next system or person can find it. At low volume that's manageable. Across thousands of documents a month arriving by email, upload, and scan, it's the reason half your document repository is filed under 'Misc' and nobody can find anything six months later.
92%
faster classification vs. manual filing
13
fields extracted & cited per document
100%
of classifications traceable to a taxonomy rule or matched template
Document in, cited filing out
Four steps, every one logged and citable back to your taxonomy.
01
Ingest
Reads the document directly from your systems
PDF, scanned image, email attachment, or office document — read from your intake channel, inbox, or upload folder. No separate conversion step.
Source document
DC-10552 — unclassified.pdf
02
Match against taxonomy
Every document checked against your taxonomy
Document type, structure, and key terms — each checked against your enterprise taxonomy and any matched templates, not a generic file-type guess.
Example query
"Does this document match the Vendor Invoice template under Finance > AP?"
03
Cite every classification
No classification without a citation
Each classification links back to the exact taxonomy node and confidence score — the citation travels with the finding, not as a separate lookup.
Cited finding
Vendor name is ambiguous between two known vendors with similar names — needs confirmation before filing.
cited to taxonomy node Finance > AP > Vendor Invoice, template match 96%04
Classify & route
One filing decision, one review queue entry
The agent returns Filed, Filed-with-confirmation, or Unclassified, with the taxonomy path and confidence score routed for review when needed.
Disposition
FiledFiled — needs confirmation ✓Unclassified
13 fields, every one cited
What the agent extracts from every document it files.
Classification summary & taxonomy path
Document type & format detection
Confidence score per classification
Key metadata extraction (dates, parties, amounts)
Ambiguous-match flag with alternatives
Duplicate-document detection
Prior classification history for similar documents
Taxonomy version & rule citation
Consolidated review queue entry
Reviewer confirmation checklist
Typical inputs
What the agent reads, and what it connects to.
Documents
- Any inbound document (PDF, scanned image, email attachment, office file)
- Your enterprise taxonomy definition
- Prior classified documents for template matching
Systems
- Document repository or content management system
- Email or intake gateway
- Enterprise search or metadata index
Is this the right fit?
Built for volume — not a one-off document.
✓ Works best for
- Teams ingesting 1,000+ documents a month across formats
- Organizations standardizing on a single enterprise taxonomy
- IT wanting every classification traceable to a rule or template match
Too small for
- A single one-off document someone can file faster than configuring the taxonomy
- Replacing a records manager's judgment on a genuinely novel document type
- Taxonomies that change faster than they can be documented
Grounded in your taxonomy, not a black box
Compliance-first, the same standard every Sphere agent is held to.
The agent classifies only against your documented taxonomy and matched templates — it flags ambiguity rather than guessing when a document doesn't clearly fit. Every classification sits between intake and filing, never past it.
Classifications reference the specific taxonomy node and confidence score directly — the same audit-trail standard every Sphere agent on the Foundry is held to.
Prerequisites taxonomy defined
A documented enterprise taxonomy is required before a classification is returned.
Human review required for ambiguous matches
A reviewer confirms any ambiguous or low-confidence classification before it's filed — never fully autonomous.
Source grounding taxonomy-cited
Classifications reference the specific taxonomy node and confidence score directly in every review.
Foundation Sphere AI Foundry
Access controls and audit trail configured here carry forward to every other agent you deploy.
Frequently asked
Does the Document Classification & Taxonomy Agent replace a records manager?
No. It handles the routine classification at intake and flags ambiguous cases — a records manager still confirms genuinely novel document types.
What document formats does it support?
PDF, scanned images, email attachments, and common office file formats out of the box. Additional formats can be added during onboarding.
How does it integrate with our document repository?
It reads directly from your document repository, email gateway, or upload folder — no separate conversion step required. Integration scope is confirmed during deployment planning.
Is every classification traceable to a taxonomy rule?
Yes. Every classification cites the specific taxonomy node and confidence score it came from — there is no classification without a citation back to the taxonomy.
How is this different from a generic OCR tool?
It's built specifically around classifying against your own enterprise taxonomy with a full audit trail, not just extracting raw text. It runs on Sphere AI Foundry, so the access controls and audit trail configured here carry forward to other agents you deploy.
Hear from
our clientsHear from our clients
Lee Ebreo
VP of Engineering at Credit Ninja
These things would not have been achievable if we did not build our own in-house system and if we did not partner with Sphere to help us achieve our goals.
Selah Ben-Haim
VP of Engineering at Prominence Advisors
Our experience with Sphere and their team has been and continues to be fantastic. We keep throwing new projects at them, and they keep knocking them out of the park (including the rescue of a project that was previously bungled by another vendor).
Ben Crawford
Senior Product Manager at Enova Financial
I would expect to be delighted. It's been a really positive experience, working with Sphere, and I would expect you to have the same.
Mark Friedgan
CEO at CreditNinja
Sphere consistently prioritizes the needs of their clients, demonstrating both agility and teamwork. As an offshore team, they have been an integral part of our organization and we plan to continue growing with them.
René Pfitzner
Co-Founder at Experify
Sphere provided excellent full-stack development manpower to augment our team and help push our product forward. They are easy to work with, tech-savvy and proactive.
Bruce Burdick
Chief Information Officer at Integra Credit
We've been working with Sphere and its excellent consultants since our founding. I've found that they are true partners in the success of our business.
Jemal Swoboda
CEO at Dabble
The resources and developers that Sphere Software provides are skilled and have the required technical expertise, but more importantly, they have helped us build a culture of excellence within our team.
Arthur Tretyak
Founder and CEO at IntegraCredit
With Sphere, we were able to migrate in half the time it would take to train an additional FTE… and for a fraction of the cost. Our experience with Sphere has been exceptional.
Lee Ebreo
VP of Engineering at Credit Ninja
These things would not have been achievable if we did not build our own in-house system and if we did not partner with Sphere to help us achieve our goals.
Selah Ben-Haim
VP of Engineering at Prominence Advisors
Our experience with Sphere and their team has been and continues to be fantastic. We keep throwing new projects at them, and they keep knocking them out of the park (including the rescue of a project that was previously bungled by another vendor).
TOP AI CODE GENERATION COMPANY UNITED STATES 2025
TOP AI TEXT GENERATION COMPANY FLORIDA 2025
TOP APP DEVELOPMENT COMPANY MANUFACTURING 2025
TOP ARTIFICIAL INTELLIGENCE COMPANY UNITED STATES 2025
TOP CHATBOT COMPANY UNITED STATES 2025
TOP RECOMMENDATION SYSTEMS COMPANY UNITED STATES 2025
GLOBAL RECOGNITION AWARDS 2026
BEST IN AI FOR DOCUMENTATION VALIDATION 2026
BEST IN AI POWERED INVOICE AUDITING 2026
GLOBAL TECH WINNER 2026
MOST INNOVATIVE TECHNOLOGY CONSULTING FIRM 2026
GLOBAL GENERATIVE AI AWARD 2026 WINNER
Start here
See it on your own document backlog
Bring a real slice of your unfiled documents to the demo. We'll show the classification live — taxonomy path, citations, and the confidence score, on your own files.
Looking for a different workflow? Browse the full Agent Catalog.
Talk to a solutions architect
Direct to a senior architect — never a sales queue. Replies within one business day.
By submitting, you agree to be contacted about this request. We never share your details.