Titulo Estágio
SaaS Access Request Automation with AI-Enhanced Governance
Áreas de especialidade
Sistemas Inteligentes
Sistemas Inteligentes
Local do Estágio
Remoto / Mindera
Enquadramento
As companies increasingly rely on third-party SaaS platforms (e.g., GitHub, Jira, Zoom, Figma, OpenAI), managing access to these tools becomes a growing operational, financial, and security challenge. Currently, most requests for access are processed manually, often lacking centralized visibility, consistent policy enforcement, or lifecycle management.
This project proposes the development of an intelligent system to streamline and partially automate how access to these tools is requested, justified, approved, and maintained. Blending rule-based logic, natural language understanding, and AI-assisted decision support.
The student will prototype a system that provides:
- A catalog of all approved SaaS platforms, including metadata like cost, owner, documentation, and risk level.
- A structured request flow that captures justification, project linkage, desired access duration, and references.
- Automated validation of requests against budgeting constraints, usage history, and company policies.
- AI-augmented justification analysis and access role recommendations (e.g., read-only vs admin).
- Access lifecycle features: periodic review reminders, expiration notifications, and de-provisioning support.
- Support for human-in-the-loop decisions where automation is insufficient.
Innovative aspects
- Provides automated, AI-augmented access governance over dozens of SaaS systems, reducing friction and human error.
- Incorporates justification understanding and role recommendations using NLP to enterprise access tools.
- Bridges multiple domains: cost control, IT operations, AI, and security policy enforcement.
- Generalizable to multiple types of SaaS tools across engineering, design, legal, HR, and operations.
Bibliography
* Saltzer, Jerome H. and Michael D. Schroeder. “The protection of information in computer systems.” Proc. IEEE 63 (1975): 1278-1308.
* ISO/IEC 27001
* Unleashing the power of AI in Decision Support Systems
Objetivo
- Build a searchable, editable SaaS tool catalog, with key information for decision-making.
- Implement a user-friendly request submission process with structured fields and free-text reasoning.
- Create a rules engine that enforces budgeting limits, department policies, and existing license quotas.
- Apply NLP to analyze and classify request justifications for decision support.
- Implement automation for access review prompts and expiration tracking.
- Provide explainability/logging for audit and ISO 27001 readiness.
Plano de Trabalhos - Semestre 1
T1.1 - Research existing access workflows and SaaS platform management strategies
T1.2 - Draft system architecture
T1.3 - Write up implementation and testing plan
Plano de Trabalhos - Semestre 2
T2 - Build tool catalog + structured request interface
T3 - Implement rules-based validation logic and NLP justification classifier
T4 - Add expiration tracking, notification system, role suggestions, and feedback loop
T5 - Document system design and results; prepare report and presentation
Condições
Work may be fully remote, partially remote or fully at office.
Laptop from Mindera
Internship allowance (food & transportation)
Orientador
Simão Belchior
simao.belchior@mindera.com 📩