// ST-DOCSEARCH-001 — DOCUMENT INTELLIGENCE
Your documents.
Every word.
One second.
Every word.
One second.
If your business runs on documents — contracts, case files, invoices, specs, medical records, policies, agreements, work orders, inspection reports — and your team wastes time every day just trying to find things, this was built for you. Law firms. Real estate offices. Medical practices. Insurance companies. Auto shops. Manufacturers. Distributors. Contractors. CEOs. If it lives in a file and you can't find it fast enough, send it to us. We make every word of every document searchable in under a second — by exact term or plain English question. No IT department required. No software to install. No packages to add on. You send the files. We handle everything else.
// THE QUESTION NOBODY CAN ANSWER FAST ENOUGH
?
Where is the bid proposal for the Phoenix library project?
?
Does this job require prevailing wage?
?
What did the Hernandez abatement spec say about tree protection?
?
Where is the Hernandez case transcript from the March deposition?
?
What was the approved quote price for 12-gauge wire on the Falcon Square job?
?
Which supplier agreement covers our copper pipe pricing for Q3?
Every one of those questions has an answer buried somewhere in a folder nobody can search fast enough. The answer exists. Finding it costs twenty minutes of someone's day — every time it gets asked. Synterminal Document Search returns the answer, the page number, and a direct link to the source document. Under one second.
// HOW IT WORKS — ST-DOCSEARCH-PIPELINE
SEND THE FILES — WE HANDLE THE REST
ST-DS-01
You send the documents.
Invoices, contracts, bid packages, specs, addenda, supplier agreements, case files, transcripts — whatever you have, in whatever condition. PDFs, scanned documents, drawings. We take them as-is.
ST-DS-02
We textify and normalize every page.
Every character of every document is extracted, cleaned, and structured. Scanned pages go through OCR. Every document gets a quality score. Every page gets a number. Nothing is skipped.
ST-DS-03
The corpus gets indexed two ways.
Full-text BM25 search finds exact terms — part numbers, clause language, names, dates. Semantic search finds meaning — you ask in plain English and the system finds the answer even when your exact words don't appear in the document.
ST-DS-04
You search. You get the answer.
Type any query. Results come back in under a second with the exact passage, the page number, and a direct link to the source PDF. The Context Pod lets you collect passages from multiple documents and paste them into any AI for instant analysis.
// TWO SEARCH MODES — ONE SYSTEM
EXACT TERMS AND PLAIN ENGLISH — BOTH WORK
§ KEYWORD SEARCH
You know the term. Find it everywhere it appears.
FTS5 full-text search with BM25 relevance ranking. Every character of every document is indexed. Boolean operators, phrase search, prefix wildcards. If the word is in the document, this finds it — with the surrounding context and the page number.
SEARCH
"prevailing wage"
SEARCH
liquidated damages AND schedule
SEARCH
WR-THHN-12-BLK
⟡ SEMANTIC SEARCH
You know the question. Find the answer.
Sentence-transformer embeddings rank documents by meaning, not just matching words. Ask in plain English. The system finds the relevant passage even when your exact phrasing doesn't appear anywhere in the document. Built for questions that don't have an obvious keyword.
SEARCH
Who is responsible for tree protection during demolition?
SEARCH
What happens if the contractor misses the completion date?
SEARCH
Where does the owner furnish materials?
// ST-VID-DOCSEARCH-001 — LIVE SYSTEM DEMO
REC
// WHO THIS IS FOR
IF YOUR ANSWERS ARE BURIED IN FILES — THIS IS FOR YOU
TRADE CONTRACTOR
You finished the job. Now someone is asking about the bid spec from two years ago. You know you have it. You just can't find it.
Bid packages, addenda, change orders, prevailing wage determinations, supplier agreements, material specs. Every document from every job, searchable in under a second. Type the part number. Type the clause. Get the page.
LAW FIRM
Discovery produced 40,000 pages. Your associate has been searching for three days. The clause is in there somewhere.
Full-text and semantic search across your entire document production. Every deposition transcript, contract, correspondence, and exhibit indexed. The clause, the page, the source file — in under a second. The Context Pod assembles the relevant passages for AI-assisted analysis.
DISTRIBUTOR
Five years of supplier agreements across twelve vendors. Nobody remembers what the copper pricing clause says and the rep isn't returning calls.
Supplier agreements, purchase orders, pricing schedules, product catalogs. All of it indexed. Search by part number, by vendor name, by material, or by plain English question. The answer is in the document. Now you can find it.
MANUFACTURER
Your compliance team needs the material certification for a component you sourced in 2022. The supplier changed names twice since then.
Material certifications, compliance documentation, spec sheets, quality records, audit reports. Indexed and searchable by any attribute in the document — including the original supplier name, the component part number, or the certification standard.
GENERAL CONTRACTOR
The owner is asking about the tree protection plan from the demolition addendum. The job was two years ago. You have twelve binders.
RFI logs, submittals, drawings, project correspondence, subcontractor agreements, inspection reports. Every document from every project indexed. Search across all of them simultaneously or filter to a single job. The answer comes back with the page number and a direct link to the PDF.
ANY BUSINESS WITH FILES
If your answers are buried in documents that nobody can search fast enough, that is the problem this solves.
Synterminal uses this system on its own internal document corpus — invoice audit files, procurement records, research papers, client engagement documents. It was built because we needed it. If you have the same problem, bring it here.
// RETAINER CLIENTS — ST-DS-RETAINER
Already on retainer?
Document search is a natural extension.
Document search is a natural extension.
If Synterminal is already auditing your invoices, running procurement intelligence, or conducting scoped research on retainer — your documents are already in the pipeline. The search layer is one step. Retainer clients get document search added to their engagement at a discounted rate because the ingestion infrastructure is already running on their corpus. The invoices being audited are the same documents being indexed. The research being conducted is the same corpus being searched. You're not buying a second product. You're unlocking a capability that already exists in your engagement.
Invoice Forensics retainer clients
Every invoice being audited is already indexed. Search across your entire invoice history by part number, supplier, date, or dollar amount.
Scoped Intelligence retainer clients
Research documents, public records, and intelligence files already in the corpus. Search across everything Synterminal has compiled on your behalf.
Custom research retainer clients
Every source document, transcript, and record produced during your engagement becomes searchable. The research compounds. So does the search.
Synterminal uses this internally
This system runs on Synterminal's own document corpus. It was built because we needed it. It works because we depend on it.
// ST-DS-POD — CONTEXT POD
Find it. Collect it.
Drop it into AI.
Drop it into AI.
The Context Pod is the feature that closes the loop. You search across your documents, find the five passages that matter, and collect them in the pod with one click. Copy everything to your clipboard. Paste into Claude, ChatGPT, or any AI assistant. Ask it to summarize the risk, draft the response, or identify the pattern. You went from a question to an AI-ready answer package in under three minutes — without opening a single folder.
01
Type any query — term, clause, name, date, or plain English question
02
Expand any result and read the surrounding document text
03
Click Add to Pod — collect up to 15 passages from across all documents
04
Click Copy All — everything goes to your clipboard with source attribution
05
Paste into any AI — get a summary, draft, risk analysis, or response in seconds
// THE DOCUMENTED COST OF NOT HAVING THIS
PRIMARY RESEARCH — CITED SOURCES
25%
Of the entire work week disappears searching for documents, information, or people needed to do the job.
Nearly half of employees surveyed said searching for what they need at work is more painful than going to the dentist.
9.3 hrs
Per week, per employee, spent searching and gathering information — not doing the work. Just finding it.
The framing is exact: hire five people, only four show up to work. The fifth is searching for answers and contributing nothing.
1 in 20
Documents gets lost. Recreating it costs 25 hours. Finding a misfiled one costs $120 in labor. Losing it costs more.
The average office spends $20 in labor to file each document. The cost of losing it is six times that — before counting the 25 hours to rebuild it.
Bring the
ugly problem.
ugly problem.
If a pricing question, data mess, supplier problem, monitoring task, or operational mystery has been sitting untouched — that is the job. Synterminal investigates what others don't have the infrastructure to find.
// OPEN A REQUEST
Direct intake for difficult information problems. Physical markets, procurement, pricing, supply chain, entity resolution, litigation support. Describe the problem. We scope the engagement.