Thesis - Human-in-the-Loop Calibrated Agent Harness

Technology, Data & Digital · Data, AI & Analytics · Artificial Intelligence · Software Engineering · UI/UX Design

In short

This thesis project explores creating an open, extensible agent harness to improve collaboration between humans and AI agents. It focuses on calibrating human involvement to tasks, moving beyond simple automation or interruption to a shared vocabulary for clarification, confidence, and decision-making.

Responsibilities

Investigate and prototype an open, extensible agent harness supporting richer human-AI agent collaboration.
Design clarification mechanisms, such as inline assumption flags and confidence indicators.
Develop an advisor component to monitor interactions and suggest clarifications.
Explore how the advisor can support both the AI agent and the human.
Implement a prototype, potentially using the PI agent harness and an external model provider.
Evaluate if a richer interaction vocabulary improves alignment, reduces rework, and maintains understanding and control.

Requirements

Knowledge in Computer Science, Human-Computer Interaction, Software Engineering, Artificial Intelligence, or a related field.
Experience in at least one programming language and an interest in building and evaluating prototypes.
Interest in large language models, AI agents, human-in-the-loop systems, or interaction design.
Eagerness to explore an interdisciplinary research problem spanning software, AI, and human factors.
Ability to formulate research questions, conduct structured evaluations, and communicate findings clearly.

Desired Qualifications

Bonus: experience with Python, TypeScript, LLM APIs, agent frameworks, or user studies.

Benefits

Outstanding opportunity to use skills and imagination to push the boundaries of what's possible.
Work on challenging projects and build solutions never seen before.
Be part of a team of diverse innovators.
Challenged, but not alone.
Join a team driven to go beyond the status quo to craft what comes next.

Skills

PythonTypeScript

#AI#Agent#Harness#Human-in-the-loop#Calibrated#Thesis#Interaction Design#Software Prototyping#Evaluation#Large Language Models

About Ericsson

Pioneering technology for connecting billions of people worldwide

Our purpose \nTo create connections that make the unimaginable possible.\n\nOur vision\nA world where limitless connectivity improves lives, redefines business and pioneers a sustainable future.\n\nOur values\nPerseverance, professionalism, respect and integrity.\n\nThe future is a place for purpose & vision – ours are clear, and we invite partners, customers and consumers to join us in our journey. \n\nFor a brighter future. For all. Let's #ImaginePossible

See all open jobs at Ericsson Read more about Ericsson Website