Carlos
Altamirano
AI Builder · Financial Analyst · MBA
Industrial Engineer and MBA with 7+ years in financial analysis across banking and professional services — now building AI systems that run in production. CleanTably is a live SaaS with a 100% pipeline success rate. FinSight Pro is a multi-agent LangGraph system that ingests SEC filings and generates investment memos via RAG. I'm the rare candidate who has restructured a $500M loan portfolio AND deployed multi-agent systems to production. Bilingual EN/ES, permanent resident.
Financial analysts spend days reading 100+ page SEC filings to extract a handful of numbers they can actually trust. Generic AI tools make it worse — they generate revenue figures and balance sheet ratios with total confidence, and they're often wrong. In finance, a hallucinated number doesn't just fail the demo. It fails the decision.
FinSight Pro solves this. A Router agent understands the question. A Doc Agent retrieves the exact passages from real SEC 10-K filings using domain-specific embeddings trained on financial text (Voyage finance-2 + BM25 hybrid search) — no guessing, only verified retrieval. A Risk Agent computes 6 key financial signals deterministically, without involving an LLM at all. Every answer cites its source: company, filing, section, date. A Synthesis Agent assembles the full due diligence brief and streams it in real time. The result: analysis that would take an analyst 4 hours, delivered in under 60 seconds — and every number is traceable.
Converts messy PDFs, invoices, and images into clean, structured Excel spreadsheets. The AI pipeline auto-classifies documents into 7 types and applies the matching extraction template. A 5-strategy JSON error handler (direct parse → markdown fences → regex → repair_json → truncation detection) achieves a 100% success rate in production. Handles batch processing of up to 10 files concurrently, per-IP rate limiting, and chunking for large PDFs. End-to-end product: backend, frontend, deployment, Google Ads, and analytics.
- Led $25M fleet appraisal — 80 Caterpillar units, team of 3 analysts, bank-grade deliverables
- 500+ individual heavy equipment valuations for banking credit decisions
- Promoted within 11 months on manager recommendation
- Designed a 3-phase AI automation prototype for the appraisal workflow — own initiative
Present
- Managed analytics and strategy for $30M portfolio of 25,000+ delinquent accounts
- Co-developed $500M COVID loan restructuring covering 150,000 accounts in 3 months
- Built SQL pipelines for daily portfolio segmentation, targeting and KPI reporting
- Promoted to Junior Manager within the same department
2021
- Managed real estate appraisal process for 100+ properties valued at $75M
- Coordinated 20+ external appraisal firms across the country
2017