Altumatim is an AI eDiscovery platform that takes a narrative-driven approach to evidence discovery — instead of keyword searches or TAR, litigation teams describe the story they want to tell and the AI identifies supporting evidence across the document set. Founded 2019 in Birmingham, MI by IP lawyer David Gaskey and data scientist Vasudeva Mahavishnu. Four modules: Storyline (auto-generates case chronologies), Topic Explorer (theme-aligned document clusters), Knowledge Browser (entity relationship mapping), and Time Machine (chronological evidence integration). EMMETT conversational assistant answers questions about the evidence corpus. Runs on Google Cloud. In a TCDI partnership case study, processed 70,000 Spanish-language documents in 70 minutes with 82% precision and 93% recall on 9 issue codes — eliminating translation costs. Fixed-fee per-matter pricing (unusual for eDiscovery, where per-GB is standard). No independent user reviews (zero G2/Capterra listings). Near-zero Reddit/community presence. No public security certifications despite handling privileged litigation ESI. ABA TechShow 2024 Startup Alley finalist. ACEDS AI education partner. Team of 2-10 employees — early-stage company with limited enterprise support capacity. Litigation eDiscovery only; not applicable to transactional, corporate, or regulatory work.
Company Info
- Founded: 2019
- Team size: 1-10 employees
- HQ: United States
- Sector: Litigation, Legal Research
What We Haven’t Verified
This page was assembled from publicly available information. Feature claims and workflow mappings are based on what the vendor and third-party listings publish — not hands-on testing or practitioner feedback.
Workflows
Based on practitioner evidence, Altumatim is used in these workflows:
What practitioners struggle with
Real frustrations from legal professionals — the problems Altumatim addresses (or should address). Sourced from practitioner reviews, Reddit threads, and case studies.
eDiscovery costs are insane — traditional vendors charge per-GB processing fees that can hit $100K+ for a single matter, making it economically impossible for small-to-mid firms to run proper discovery
500K documents to review, contract attorneys burning out after 4 hours of screen-staring, nobody knows if the review is consistent across 20 reviewers — and the partner watching the budget bleed
Litigation team building a case chronology across 50,000 documents, 30 depositions, and hundreds of exhibits does it in Excel or Word — no single platform connects facts, people, events, and evidence into a searchable timeline, so critical connections between a witness statement and a document are missed
MDL or mass tort coordinating counsel needs to manage document exchange among 40+ plaintiff firms, defense counsel, and the transferee court — filings need to reach all parties simultaneously, the court wants a single organized case file, and the Special Master is demanding a reliable system for tracking what was served to whom and when, but email chains with 200 attorneys are unmanageable and PACER alone does not handle the volume of inter-party communications
Cross-border litigation generates 100,000+ documents in 3-5 languages — the firm either pays $200K+ for human translation or reviews only the English documents and hopes nothing critical was in the foreign-language set, which is a malpractice risk the partner loses sleep over.
Where it fits in your workflow
Before Altumatim
Data collection (Pinpoint Labs integration) → document processing → evidence ingestion into altumatimOS. No documented Relativity export/import compatibility.
After Altumatim
Evidence identified → exported for use in motions, depositions, trial prep (document-drafting-automation) → case presentation. Export formats not documented.
Integrations & hand-offs
Pinpoint Labs for data collection. TCDI for managed review services and litigation support. Google Cloud for infrastructure. No documented DMS integrations (iManage, NetDocuments). No API for custom integrations. Data flow between Altumatim and other eDiscovery tools is opaque.
Community Data
Loading practitioner-sourced data…