Proprietary Software — Not Open Source. This repository is a portfolio showcase. Source code is not included. All rights reserved. © 2025–2026 Ahmad Al Baba.
DepositIQ is a full-stack AI legal intelligence platform that transforms raw deposition transcripts and recordings into structured, trial-ready analysis reports — in under two minutes. It is designed for litigation attorneys, legal teams, and law firms who need to move from raw testimony to actionable courtroom strategy without spending days manually reviewing transcripts.
Live Product: depositionai-5xx7rbra.manus.space
Deposition review is one of the most time-intensive tasks in civil litigation. A single deposition transcript can run 200–400 pages. Attorneys must manually cross-reference testimony for internal contradictions, assess witness credibility, extract damaging admissions, and build impeachment packages — often under tight pre-trial deadlines. This work is expensive, error-prone, and does not scale.
DepositIQ ingests deposition transcripts (PDF, DOCX, audio, video) and applies a multi-stage AI pipeline to produce:
- Contradiction detection — flags statements that contradict each other across the same or multiple depositions, with severity scoring
- Witness credibility scoring — a 0–100 credibility index derived from linguistic patterns, evasiveness, and internal consistency
- Key admissions extraction — surfaces statements that constitute legal admissions, categorized by topic
- Sentiment and tone analysis — tracks emotional shifts across the deposition timeline
- Argument drafting — generates trial-ready argument outlines grounded in the deposition record
- Impeachment packages — builds ready-to-use impeachment sequences for cross-examination
- Case timeline — reconstructs a chronological narrative from all depositions in a case
- Cross-witness comparison — identifies conflicts between multiple witnesses on the same facts
- Live deposition mode — real-time AI assistance during active depositions
- AI chat — ask natural language questions about any deposition or case
| Feature | Description |
|---|---|
| Multi-format ingestion | PDF, DOCX, MP3, MP4, YouTube URL, Google Drive link |
| OCR support | Scanned PDFs processed via Tesseract OCR |
| Multi-deposition cases | Organize unlimited depositions under a single case |
| Contradiction heatmap | Visual matrix showing conflict density across witnesses |
| Witness profiles | Per-witness credibility history and scoring trends |
| Prep outlines | AI-generated deposition preparation checklists |
| Bulk export | Export full case analysis packages as PDF reports |
| Team collaboration | Share cases with colleagues via secure share tokens |
| Annotations | Highlight and annotate transcript passages |
| Follow-up questions | AI-suggested follow-up questions based on testimony gaps |
| Case insights | Aggregate intelligence across all depositions in a case |
| Live product demo | Interactive in-page walkthrough on the landing page |
| Stripe billing | Free tier (3 depositions) + Solo, Firm, Enterprise, Founders plans |
| AppSumo integration | Lifetime deal redemption and code stacking |
See the
SCREENSHOTS/directory for a full capture guide.
| Screen | Description |
|---|---|
| Landing page | Hero, interactive demo, features, pricing |
| Dashboard | Case list with status indicators |
| Case page | Toolbar with all analysis tools embedded inline |
| Deposition analysis | Overview tab with ScoreRing, credibility, admissions |
| Contradiction heatmap | Cross-witness conflict matrix |
| Argument draft | AI-generated trial argument outline |
| Upload page | Persistent upload zone with URL import bar |
See TECH_STACK.md for the full rationale behind each choice.
| Layer | Technology |
|---|---|
| Frontend | React 19, Tailwind CSS 4, shadcn/ui, Radix UI |
| API layer | tRPC 11, Zod, Superjson |
| Backend | Node.js, Express 4 |
| Database | MySQL / TiDB (Drizzle ORM) |
| AI | LLM via Manus Forge API |
| Storage | AWS S3 |
| Auth | Manus OAuth 2.0 |
| Payments | Stripe |
| Transcription | Whisper API |
| OCR | Tesseract |
| Video ingestion | ytdl-core (YouTube), Google Drive API |
See ARCHITECTURE.md for the high-level system diagram.
See PRODUCT_BRIEF.md for the investor-facing one-pager.
See DEMO_GUIDE.md for the recommended 3–5 minute walkthrough script.
See METRICS.md for performance benchmarks and milestone tracking.
See LICENSE. All rights reserved.
Ahmad Al Baba — Solutionist & Advisor For access inquiries, partnership, or licensing: contact via the live product at depositionai-5xx7rbra.manus.space
Built with DepositIQ — Ahmad Al Baba — 2026-05-12