Agentic AI Assurance & Runtime Governance
Evidence-conditioned authority, bounded autonomy, escalation, abstention, human approval and auditable runtime decisions.
Explore research →Yasir Siddiq · Barcelona, Spain
Software Engineering & Trustworthy Agentic AI
Research and engineering work on AI assurance, runtime governance, multi-agent reliability, failure containment and bounded operational authority, with applications to software systems and increasingly autonomous 5G/6G networks.
Trustworthy Agentic AIevidence · authority · reliability
01AI Assurance & Runtime Governance
02Multi-Agent Reliability
03Software Engineering for AI
04AI-Assisted 5G/6G
Research focus
Problems I am investigating
Multi-Agent Reliability & Failure Containment
Failure propagation, intermediate-output verification, context isolation, containment, recovery and workflow observability.
Explore research →Software Engineering for AI Systems
Reproducible evaluation, explicit contracts, testing, traceability, policy-as-code and dependable engineering of AI-assisted systems.
Explore research →AI-Assisted 5G/6G Systems
AI-assisted network management, fault classification, runtime assurance and bounded operational authority for increasingly autonomous telecom systems.
Explore research →Selected research
Flagship technical work
Research Prototype · 2026
Policy-Gated Authority for AI Agents in DevSecOps
A deterministic technical-governance prototype investigating when an AI software-engineering agent should be permitted to move from proposing an action to performing a consequential DevSecOps action.
Research Demonstrator · 2026
Agentic AI Resilience Lab
A bounded, deterministic simulation environment for studying failure propagation, evidence checking, containment, escalation, context isolation and recovery in multi-stage agentic workflows.
Research Prototype · 2026
GATE-5G — From Prediction to Permission
A technical-governance research prototype separating AI inference from operational authority in AI-assisted 5G/6G network management, connected to preserved fault-classification evidence.
Selected outputs
Publications & policy outputs
- Large Language Models in AI-Augmented DevSecOps Pipelines: Exploring Secure Software Engineering Practices and Architectural ChallengesInternational Journal of Advanced Research · Vol. 3 · Issue 3 · 2026Published 31 July 2026.
- A Leakage-Safe Study of Trust-Aware Fault Classification in ns-3/5G-LENA Using QoS-Derived FeaturesWECE 2026 · 2026Accepted for WECE 2026; publication and publisher citation pending.
- Before AI Can Merge or Deploy: Minimum Governance Controls for Autonomous Agents in DevSecOps PipelinesZenodo · 2026Public policy and technical-governance output; Zenodo deposit does not constitute peer-review validation.
- GATE-5G: From Prediction to Permission in AI-Assisted 5G NetworksZenodo · 2026Policy memo presenting a proposed technical-governance framework; not a formal standard or independent validation.
Selected engineering
Controlled AI automation in practice
Engineering MVP · 2026
AI Lead Qualification & Human-Reviewed Follow-Up Automation
A functional n8n workflow MVP in which an LLM extracts structured lead information, deterministic rules score and route the lead, and Gmail prepares a draft for human review rather than sending automatically.
Control pattern
- 01LLM interprets
- 02Deterministic rules decide
- 03Automation prepares
- 04Human authorizes
Background
Software engineering, telecommunications and teaching
My background combines an MS in Software Engineering, telecommunications engineering experience with Alcatel-Lucent, Huawei and SOMTEL, university teaching, and more recent work on reproducible AI/ML evaluation and technical AI assurance.
2006
BS Software Engineering
UET Taxila
2007–2015
Telecommunications Engineering
Alcatel-Lucent · SOMTEL · Huawei
2017–2023
University Teaching
Computer Science / Software Engineering
2026 →
Trustworthy Agentic AI
AI assurance · runtime governance · multi-agent reliability