I'm a full-stack engineer. Since 2020 I have built and kept in production Skuld, an investigation platform for fraud, sanctions and compliance teams: one search queries 15 sources, a graph connects what they return, and an AI assistant reads it. I work across the whole delivery: the interface, the APIs, the integrations, the tests and what breaks in production.
See the live demos at davidvalor.vercel.app · Download my CV
The products belong to Grupo Analytiko and their code is private. The demos on my site rebuild how they work with fictional data.
| Repository | What it does |
|---|---|
| opensanctions-api-free | Sanctions screening on OpenSanctions open data. Loads it into MongoDB and serves a search API that ignores accents and case, because a misspelled name must not let a sanctioned person through. Doubtful matches go to a model that writes down its reason for the audit trail. Node.js MongoDB AWS Lambda |
| gacetas-scrapper | Downloads the PDFs of Venezuela's Official Gazette archive, extracts the text page by page with OCR and indexes it, so a search by name finds appointments, companies and penalties published years ago. Python OCR |
| texas-solver | A game-theory solver for Texas Hold'em. Python |
Backend TypeScript, JavaScript, Python, Node.js, Express, NestJS, REST, GraphQL, Socket.IO
Frontend and data Angular, React, Vue, Next.js, MongoDB, Neo4j, PostgreSQL, SQL Server
AI OpenAI and Anthropic APIs, LangChain, RAG, agents, human-in-the-loop review, token-cost controls
Cloud and testing Google Cloud Run, AWS (Lambda, AppSync, S3), Docker, CI/CD, Playwright
Domain OSINT, sanctions and PEP screening, KYC and third-party due diligence, adverse media, transaction-graph analysis
Available for remote roles and projects. Venezuela, GMT-4.