Welcome to AgenticDev 2026 (ACM proceedings) Panel Discussion: The AGenSE Alliance: Shaping the Future of Agentic Software Engineering The panel members present the Horizon Europe’s Vision for Agentic Software Engineering with five projects AI4SWEng, ASTIR, HIVEMIND, MOSAICO, and FAITH.
AI4SWENG - Ali Atalay, AI4SEC, Estonia
ASTIR - Dr. Anh Tuan Le, University of Warwick, UK
FAITH - Ioannis Kyprakis, Foundation for Research and Technology (FORTH), Greece
MOSAICO - Dr. Jean-Marie Mottu, Nantes Université, France, and Dr. Tuong Manh Vu, University of York, UK
Context Inheritance: A Git-Native Architecture and Pre-Registered Study of Repository Context for AI Coding Agents (#22) - Pavel Kerbel, Milana Kerbel, Vitali Abramov, Liat Abramov (Metatron Research)
10:50–11:10
Spec-Graph Meta-Framework for AI-Assisted Software Development: An Industry Case (#86) – Daniel Soto (Tres Astronautas), Daniel González-Rueda, Mario Linares-Vásquez (Universidad de los Andes), Andrés Hurtado (Blend 360)
11:10–11:20
Extender: A Multi-Agent System for Code Extensibility Review (#21) – Jiahe (Jay) Chen (Duke Kunshan University), Jiacheng Shen (Duke Kunshan University)
Session 2: Agentic Coding
11:30–11:50
BDD-Bench: A Behavior-Driven Benchmark for Coding Agents Across the Application Lifecycle (#25) – Michał Jureczka (Jagiellonian University in Krakow; Exadel), Anna Paleczek (AGH University of Krakow; Exadel), Aiden Campbell (Exadel), Karol Przystalski (Jagiellonian University in Krakow; Exadel)
11:50–12:10
Towards Model-Driven Generation of Domain-Specific Modelling Agents for Local Language Models (#58) – Antonio Garcia-Dominguez, Ionut Predoaia, Tuong Manh Vu, Konstantinos Barmpis, Dimitris Kolovos (University of York)
12:10–12:20
Defining and Deploying Agentic Swarms with the BESSER Low-Code Platform (#102) – Adem Ait (University of Luxembourg), Javier Luis Canovas Izquierdo (Universitat Oberta de Catalunya), Jordi Cabot (Luxembourg Institute of Science and Technology & University of Luxembourg)
12:20–12:30
Value-Preserving Architectures for Agentic AI Systems (#100) – Alessandro Pesare, Tommaso Dolci, Katja Hose, Emanuel Sallinger (TU Wien)
An Evidence-Preserving Architecture for LLM-Assisted Vulnerability Review of Robotics Software (#99) – Anhtuan Le, Gregory Epiphaniou, Carsten Maple (WMG, University of Warwick)
14:40–15:00
Teach-to-Crash: A Closed-Loop Student–Teacher LLM Framework for Collision-Inducing Test Scenario Generation (#126) – Zaid Ghazal (University of Michigan-Dearborn), Khouloud Gaaloul (University of Michigan Dearborn), Bruce Maxim (University of Michigan)
Session 4: Repair
15:20–15:40
Before the First Model Call: Factorizing Token Alignment and Cue Weighting for Coding-Agent Routing (#117) – Nikhil Mahesh (Independent Researcher, USA)
15:40–15:50
Schema-Conditioned Next-Tool Prediction for Coding Agents Under Interface Drift (#75) – Selma Nyberg, Isak Samsten (Department of Computer and Systems Sciences, Stockholm University)
15:50–16:00
Who Should Own the Loop? Harness Decomposition for Small-Model Repository Repair (#131) – Francesco Dente (EURECOM), Dario Satriani, Donatello Santoro, Enzo Veltri (University of Basilicata), Paolo Papotti (EURECOM)
Toward Automated Search for Verifier-Spending Agentic Workflows: A Vision, Preliminary Evidence, and the Obstacles in the Way (#134) – Andrey Spiridonov (AutoAW)
16:40-16:50
MOSAICO: A Multi-Agent Infrastructure for Collaborative LLM-Powered Software Engineering (#159) – Tuong Manh Vu, Ionut Predoaia, Konstantinos Barmpis, Dimitris Kolovos, Antonio Garcia-Dominguez (University of York)
16:50–17:30
Closing Remarks and Poster Session
Accepted Posters
katIA: Hybrid Source Code Change Impact Analysis - Katharina Stengg (University of Klagenfurt), Christian Macho (University of Klagenfurt), Martin Pinzger (University of Klagenfurt)
Evaluating Low-Shot Procedural Skill Transfer in Language-Agent Debugging - Jiachen Sun (ShanghaiTech University)
Evidence-Driven, Failure-Resilient Planning for Domain-Agnostic Multi-Agent Orchestration - Baran Osmanoglu (Saura AI)
Agents as Architecture Evaluators: Automating ATAM and SAAM for Continuous Assessment - Adyansh Kakran (International Institute of Information Technology, Hyderabad), Sathvika Miryala (International Institute of Information Technology, Hyderabad), Karthik Vaidyanathan (International Institute of Information Technology, Hyderabad)
Automated Annotation of Test Types Using LLMs: A Multi-LLM Workflow for Machine Learning Test-Related Commits - Asli Sari (Istanbul Technical University), Ayse Tosun (Istanbul Technical University), Gulfem Isiklar Alptekin (Galatasaray University)
Documentation Retrieval by AI Coding Agents: An Empirical Study of HTTP Behavioral Signatures - Oleksii Borysenko (Cisco DevNet)
Execution-First Synthetic Tool-Use Trace Generation for LLM Agents - Hafsa Ouajdi (EURECOM Sophia Antipolis, France Aily Labs Munich, Germany), Francesco Giannuzzo (EURECOM Sophia Antipolis, France Aily Labs), Alaa Boukhary (Aily Labs Munich, Germany), Paolo Papotti (EURECOM Sophia Antipolis, France), Gerard Conangla (Aily Labs Munich, Germany), Adam Elwood (Aily Labs Munich, Germany)
Other accepted contributions (not-in-person presentations)
They Do Not Fire Late, They Barely Fire (#195) – Simarjot Khanna
Cost-Effective Repository Exploration for Agentic Issue Localization (#18) – Mohammad Nour Al Awad, Sergey Ivanov (ITMO University St. Petersburg), Russia
Measuring the Wrong Number: Cost-Aware, Correctness-Gated Evaluation of Efficiency Claims in Multi-Turn Agentic Coding (#20) – Erdni Zergetaev
Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair (#114) – Xueping Gao, Jianwei Yang, Qiang Yang (Alibaba Cloud Computing)
Get Calendar (iCal)
iCalendar subscription service for your personal schedule
You first need to have starred events () in order to use the iCalendar subscription service.
Currently Viewed Program
Downloads the currently viewed program (with filters applied) in iCal format