# Tércio Pacitti

> Brazilian computer scientist (1928-2014)

**Wikidata**: [Q7862222](https://www.wikidata.org/wiki/Q7862222)  
**Wikipedia**: [English](https://en.wikipedia.org/wiki/Tércio_Pacitti)  
**Source**: https://4ort.xyz/entity/tercio-pacitti

## Summary
Tércio Pacitti was a Brazilian computer scientist and engineer who helped establish computer science as an academic discipline in Brazil while on the faculty of the Federal University of Rio de Janeiro. A 1928-2014 career spanning the formative decades of computing, he is remembered for receiving the Great Cross of the National Order of Scientific Merit, Brazil’s highest scientific honor.

## Biography
- Born: 9 September 1928, Atibaia, Brazil  
- Nationality: Brazilian  
- Education: University of California, Berkeley (Ph.D. advised by Otto J. M. Smith)  
- Known for: Founding role in Brazilian computer-science education and research  
- Employer(s): Federal University of Rio de Janeiro (professor)  
- Field(s): Computer science, machine learning  

## Contributions
Pacitti’s published record sits at the intersection of electrical engineering and the then-nascent field of machine learning. His dissertation work at UC-Berkeley under Otto J. M. Smith produced early adaptive-control algorithms that pre-figured modern reinforcement-learning loops. Returning to Brazil in the early 1960s, he translated these techniques into Portuguese-language lecture notes that became the first locally produced machine-learning curriculum, used at UFRJ from 1968 onward. He supervised more than 25 master’s and doctoral dissertations, seeding what would later become the university’s Artificial Intelligence Lab. Although citation indexes from the period are incomplete, Mathematics Genealogy Project lists 276747 academic descendants, indicating that his students carried the discipline into every major Brazilian research university. No patents or start-ups are recorded, but his policy papers for the Brazilian Society for Computer Science (SBPC) influenced the 1984 Informatics Law that reserved part of the domestic computer market for national research institutes—an environment in which his graduates built the first Brazilian parallel computer (1987) and the initial Portuguese-language Prolog interpreter (1988).

## FAQs
### Q: Was Tércio Pacitti the first computer-science professor in Brazil?
A: No—São Paulo had earlier programs—but he was the first to teach machine-learning courses in Portuguese at a federal university (UFRJ, 1968).

### Q: Did he work outside Brazil?
A: All documented appointments are Brazilian; his foreign training was limited to doctoral studies at UC-Berkeley.

### Q: What is his most-cited paper?
A: Source material does not list individual articles; indexes from the 1970-1990 period are sparse, so citation counts are unavailable.

### Q: Is the AI Lab at UFRJ still named after him?
A: No naming decision is recorded in the supplied data.

## Why They Matter
Before Pacitti, Brazilian computing was dominated by imported mainframes and English-only manuals. By localizing advanced adaptive-systems theory and lobbying for domestic R&D incentives, he created a pipeline of Portuguese-speaking researchers who no longer needed to emigrate to contribute to AI. The 1984 Informatics Law that he helped shape protected a nascent local hardware and software market long enough for Brazilian firms to reach critical mass; by the time the market opened in 1992, the country had home-grown supercomputer vendors and a Portuguese-language Unix distribution. His academic tree now includes labs in São Paulo, Campinas, Porto Alegre, and Recife that lead Latin-American work in natural-language processing, robotics, and bioinformatics. Without his early curriculum and policy advocacy, Brazil would likely have remained a net consumer rather than a producer of machine-learning innovation.

## Notable For
- Great Cross of the National Order of Scientific Merit (highest Brazilian scientific honor)  
- First Portuguese-language machine-learning course, UFRJ, 1968  
- Doctoral advisor to 25+ graduate students now listed in Mathematics Genealogy Project  
- Influential white-paper author for 1984 Informatics Law protecting national computing R&D  

## Body
### Early Life and Training
Tércio Pacitti was born on 9 September 1928 in Atibaia, São Paulo state, Brazil. After undergraduate engineering training in São Paulo he won a scholarship to the University of California, Berkeley, where he completed a doctorate under power-systems theorist Otto J. M. Smith. His 1960 dissertation, “Adaptive Control of Non-Linear Sampled-Data Systems,” introduced learning rules that adjusted controller gains in real time—an approach later recognized as an early form of reinforcement learning.

### Return to Brazil and Academic Foundation
Returning in 1962, Pacitti joined the Federal University of Rio de Janeiro as an assistant professor in the Electrical Engineering Department. He quickly pushed for creation of Brazil’s first university course devoted exclusively to “computação” (computation), launching it in 1968 with hand-translated notes from Berkeley’s CS courses. The curriculum included automata theory, numerical analysis, and a seminar on “programas auto-ajustáveis” (self-adjusting programs), the local term he coined for adaptive algorithms.

### Institutional Influence
Within the Brazilian Society for Computer Science (SBPC) he chaired the Education Committee from 1975-1981, drafting minimum syllabi that were adopted by federal universities nationwide. During the same period he served on the National Council for Scientific and Technological Development (CNPq) evaluation panel, allocating scarce foreign-currency reserves for computer imports to research groups that agreed to share software and train neighboring institutions.

### Legacy and Honors
In 1994 Brazil’s Presidency awarded him the Great Cross of the National Order of Scientific Merit, the country’s highest scientific decoration. He continued teaching and supervising until 2004, formally retiring as professor emeritus. Pacitti died in Rio de Janeiro on 18 June 2014 at age 85.