Capital Origination Engine
A Python-based issuer scoring and outbound automation system that converts SEC filings, market data, and financial signals into prioritized financing pipeline.
I built an AI-assisted capital markets GTM engine that screens small-cap and micro-cap public companies, detects financing signals from SEC filings, scores issuers across five qualification dimensions, generates evidence-backed outreach, and exports campaign-ready workbooks for outbound execution.
The system connects public-company research to sales execution by turning market data, EDGAR filings, XBRL financials, financing disclosures, and stakeholder information into ranked issuer outputs and personalized outreach workflows.
Public-company financing origination is slow, manual, and hard to scale.
Before contacting an issuer, a capital markets or sales team needs to review filings, financing history, liquidity, capital structure, trading profile, shareholder approval risk, and relevant executive stakeholders. This creates several bottlenecks:
- High research time per issuer
- Inconsistent qualification criteria
- Generic outbound messaging
- Poor prioritization across large issuer universes
- Long sales cycles with weak follow-up infrastructure
A 9-step Python-based origination engine.
The engine converts public-company data into a ranked, evidence-backed, campaign-ready output. It combines public market data, SEC filings, financial statement data, signal parsing, issuer scoring, stakeholder enrichment, and outbound message generation.
Load and Validate Ticker Universe
The engine begins by loading a ticker universe from command-line arguments, text files, or a sample ticker list. The system validates, normalizes, and deduplicates tickers before downstream enrichment begins.
Enrich Market Data
Each issuer is enriched with market data, including share price, market capitalization, average daily trading value, float, and shares outstanding. This helps evaluate potential financing feasibility, trading liquidity, and estimated draw capacity.
Resolve CIKs Through EDGAR
The engine resolves each ticker to its corresponding CIK through EDGAR. This step allows the system to retrieve issuer-specific SEC filings programmatically.
Retrieve SEC Filings
The engine retrieves relevant SEC filings, including capital-structure filings, periodic reports, and event filings. Relevant filing types may include S-3, S-1, F-3, 424B5, 10-K, 10-Q, 8-K, and 6-K.
Parse Financing Signals
The system parses filing text to identify financing-related signals. Detected signals may include ATM facility, equity line, shelf registration, going concern language, reverse split, PIPE activity, repeated financing activity, large shelf ratio, and active facility language. This step converts dense filing text into structured issuer-level signals.
Score Issuers Across Five Dimensions
The scoring engine evaluates issuers across five dimensions: capital need, structure fit, execution profile, outreach potential, and confidence. The system also computes estimated draw capacity and assigns each issuer a priority tier and suitability bucket.
Estimate Financial Urgency from XBRL
The engine uses EDGAR XBRL companyfacts data to estimate financial position. The system extracts cash balance and operating cash flow where available, then computes estimated runway in months. Financial urgency is classified into categories such as Critical, Near-Term, Moderate, and Low.
Generate Outreach Messaging
The outreach layer selects an account-specific angle and generates messaging inputs. The system produces a primary outreach angle, suggested subject line, suggested opener, talking points, caution notes, recommended channel, email copy, and LinkedIn message copy.
Export Excel and CSV Files
The engine exports a multi-tab workbook with structured results. The workbook includes Ranked Results, Financing Flags, Evidence, and Outreach Messaging — designed to support account review, rep preparation, campaign execution, and pipeline tracking.
A multi-tab, campaign-ready workbook.
The engine produces a multi-tab workbook with:
Sensitive contact information, issuer pipeline notes, and non-public transaction details are redacted from public-facing examples.
A full GTM intelligence workflow.
This is not just a lead list or filing parser. The system creates a full go-to-market intelligence workflow:
Public data → signal extraction → issuer scoring → evidence layer → stakeholder mapping → personalized outreach → campaign execution
That matters because complex financing products often have 12–24+ month sales cycles. Account prioritization, evidence-backed outreach, and disciplined follow-up are critical.
Designed and built end-to-end.
I designed and built the workflow end-to-end, including:
Hands-on experience across the stack.
Selected issuer names, contact information, financing counterparties, and pipeline details have been redacted. This case study is for portfolio demonstration only and does not constitute investment, financing, securities, legal, or financial advice.
Built by Bryce Clark as a portfolio project demonstrating AI-assisted GTM systems, capital markets research, and outbound pipeline development.