Junior Analyst
Pedro Francisco arrived at energy analysis through engineering. Years spent building systems that move, reconcile, and validate data taught him that the hardest problems in any data pipeline are rarely computational. They are problems of definition. A system can process millions of records flawlessly and still produce meaningless output if the underlying labels do not mean what the analyst assumes they mean.
That lesson became concrete when he began working with public energy filings. The same financial line appears under different names across companies, and sometimes across consecutive quarters of the same company. One filing reports total revenues; another separates the same figure into components that must be summed. A balance sheet item classified as a current asset in one company's presentation sits under long-term assets in another. None of this is error on the part of the filers, and all of it is invisible until someone tries to compare five companies side by side. Analysts who should be interpreting numbers spend their time reconciling labels instead.
Pedro's response to that problem is a conviction that shapes how he works: every figure must be traceable to the specific line of the specific filing that produced it. Standardized naming is only trustworthy when the mapping between a company's language and the common vocabulary is explicit, documented, and reversible. Data quality is not preparation for analysis. It is the first act of analysis, and the one that determines whether everything downstream can be believed.
Applied to the work, this means building extraction that pulls income statements and balance sheets out of 10-K and 10-Q documents and delivers them as structured, labeled spreadsheets in seconds rather than hours—while preserving the link back to the source at every step. It means treating the naming dictionary as shared infrastructure rather than as a set of rules improvised company by company. It means validating output against the original document, not against the pipeline's own expectations, because a system that fails silently is more dangerous than one that fails loudly. His technical foundation spans Python, SQL, relational database design, and Windows Server environments.
At Permian Ventures, Pedro works as a junior analyst supporting the data layer that the platform's comparative analysis depends on. The standardized naming conventions, the consistent unit handling, the period alignment that lets one company be measured honestly against another—these are the quiet mechanics beneath every chart and every insight the platform surfaces. He is drawn to that work precisely because it is foundational. Rigorous financial intelligence begins long before the first conclusion is drawn. It begins with getting the numbers right.