Human–AI Hybrid Delivery Isn’t Optional—A Robotics Product Leader Explains Why

I’m passionate about bridging technical execution with ethical and strategic leadership — helping teams deliver impactful products while advancing responsible innovation in emerging technologies.

Delivery rituals are less about manual oversight and more about interpretation, judgment, and strategic guidance. AI acts as a copilot — handling analysis and pattern recognition — while humans drive vision, ethics, and collaboration.

Share This Quote on:

• Share on Twitter
• Share on LinkedIn
• Share on Facebook

Kateryna Portmann

Senior Product Manager at ANYbotics

How AI is reshaping product management and project delivery

My role and approach to AI in project delivery are evolving significantly. Tasks that were previously manual or focused on tracking and reporting are increasingly automated, which means I spend less time on administrative work like status updates, basic task coordination, or monitoring routine KPIs.

At the same time, I’m spending more attention on areas where human judgment and AI in strategic planning add the most value: defining product strategy, integrating AI responsibly into robotics systems, anticipating regulatory and ethical considerations, and guiding cross-functional teams to align on innovative solutions.

More broadly, project delivery now requires a stronger emphasis on data-informed Ai in decision making, AI ethics, and adaptive planning. Teams are expected to iterate faster, leverage AI-driven insights for prioritization, and respond proactively to evolving technological and regulatory landscapes. My role is shifting from primarily managing tasks to orchestrating innovation responsibly and ensuring AI initiatives create real value without compromising safety, compliance, or trust.

Why human-in-the-loop systems are critical for AI-driven project delivery

Early on, I expected AI to mainly automate routine work — like tracking project progress or generating reports — but what I found is that it really excels at highlighting patterns, predicting risks, and surfacing insights that guide smarter decisions.

For example, in the Sandbox AI project, AI tools didn’t just help us analyze data faster — they revealed subtle compliance and operational risks we might have missed otherwise. That forced us to rethink how we structure workflows, prioritize tasks, and engage stakeholders.

Overall, the aspects of delivery that feel most ripe for support are data-intensive, repetitive, or predictive AI in project task automation. That means:

• Progress tracking and reporting: AI can automatically consolidate updates from multiple teams, highlight bottlenecks, and generate dashboards in real time.
• Risk identification and scenario planning: Machine learning models play a part in AI risk management. They can analyze project data to predict potential delays, resource conflicts, or compliance risks before they materialize.
• Prioritization and resource allocation: AI in resource management can suggest optimal sequencing of tasks or redistribution of resources based on historical data and project constraints.

Implementation typically involves integrating AI-powered project analytics into project management tools like Jira, Asana, or custom internal systems. For predictive modeling, we could use lightweight machine-learning pipelines that process historical project data and continuously improve with new inputs.

However, human judgment is still indispensable in several areas:

• Ethical and regulatory decision-making: Determining what is permissible, responsible, and aligned with AI in compliance and broader organizational values.
• Stakeholder communication and negotiation: AI in stakeholder management can suggest talking points, but nuanced conversations and consensus-building require empathy and persuasion.
• Strategic planning and innovation: Humans must interpret AI insights in context, decide on trade-offs, and shape long-term vision.

In short, human judgment is still necessary. It's crucial. AI in project management is a powerful force multiplier for efficiency and insight, but it works best when humans retain control over complex, strategic, and value-driven decisions.

kateryna's tip

Human judgment is still necessary. It’s crucial. AI is a powerful force multiplier for efficiency and insight, but it works best when humans retain control over complex, strategic, and value-driven decisions.

Why lightweight delivery systems don’t always work

It's important to note that, in hardware and robotics, moving away from traditional project management methods is not always easy — physical systems have long lead times, strict dependencies, and safety or compliance constraints that require careful planning. Iterative flexibility is more limited compared to purely software projects.

That said, in initiatives like Women in Robotics, we’ve been able to experiment with more lightweight, agile approaches. For example, we use Slack — and its AI features — for quick coordination across volunteers and participants. This allows us to iterate on programs, gather feedback, and adjust initiatives much faster than a traditional top-down planning approach.

So, while hardware projects often demand structured processes, lightweight systems can thrive in community-driven or cross-functional initiatives, enabling faster adaptation, greater engagement, and continuous improvement.

A real-world example of modern delivery systems navigating complexity

Here's a good example of AI delivery in practice. With the Sandbox AI project, we navigated a highly complex environment involving multiple stakeholders, regulatory constraints, and cutting-edge AI systems. The project aimed to explore safe, compliant, and innovative AI deployments in public-sector applications, which required balancing speed of delivery with governance and ethics.

Setup and approach:

• We established a cross-functional core team including engineers, policy experts, legal advisors, and AI ethicists.
• We used Agile delivery systems, leveraging Jira for sprint planning, Confluence for documentation, and Miro for collaborative design workshops.
• AI-driven tools were integrated for risk analysis and scenario simulations, allowing us to model outcomes before deploying AI-driven processes in real-world settings.

Execution:

• We started with small-scale prototypes to test compliance and ethical frameworks, iterating weekly based on both technical performance and regulatory feedback.
• Regular stakeholder review sessions ensured alignment with governance standards, while AI analytics highlighted potential compliance risks before they materialized.

Results:

• Evaluated readiness for ISO 42001
• Lessons learned fed directly into the AI House Davos 2025 white paper, where we distilled eleven key governance elements for broader industry adoption.

This project highlighted how modern delivery systems and AI-driven tools can manage complexity, accelerate iteration, and maintain ethical and regulatory compliance — even in highly uncertain environments. It reinforced the value of blending structured delivery practices with AI insights to navigate challenging projects successfully.

A tech stack that amplifies collaboration and reduces friction

I’m currently on maternity leave, so I haven’t been hands-on with day-to-day project delivery over the past months — and AI moves so quickly that it would be unfair to speak as though I have direct experience with all the latest tool decisions at ANYbotics.

That said, through my work with Women in Robotics, I’ve continued to stay engaged with trends in project delivery and collaboration. In that context, the tech stack often revolves around collaboration and knowledge-sharing tools:

• Slack for communication and coordination: This AI collaboration tool helps keep global teams connected and reduce email overload. I use Slack’s built-in AI for summarizing long threads and channels, which is especially helpful when I’m joining ongoing conversations or syncing across distributed teams. I also use it to draft quick responses or rephrase updates to make them more clear. The impact has mainly been faster context-switching and more consistent communication across the team.
• Figma for visual collaboration, particularly for workshops, brainstorming, and co-design sessions: With Figma’s AI features, I generate initial UI variations, explore layout options, and automatically document design components. It speeds up early-stage ideation and lets me share visual direction with stakeholders much earlier in the process. But the AI-generated variants aren’t final — they help jump-start design discussions and reduce iteration time.
• Notion for documentation, tracking initiatives, and making insights easily accessible to all members: Notion AI is part of my daily workflow. I use this AI task manager for drafting project documentation, meeting notes, summarizing long research docs or interviews, converting rough brainstorms into structured plans or action items, and rewriting content in different tones depending on the audience. The biggest impact is time savings — turning unstructured information into something clear and shareable takes a fraction of the time it used to.

Over the past year, I’ve noticed teams are doubling down on tools that facilitate asynchronous collaboration and transparency, particularly as hybrid work becomes the norm. Tools like Notion have proven invaluable because they allow members to contribute and review work regardless of time zone. Conversely, some legacy task-tracking tools that were rigid or siloed have been de-emphasized, as they don’t support flexible workflows or rapid iteration.

Overall, my assessment is that workflow automation software succeed when they amplify collaboration, reduce friction, and enable teams to focus on impact rather than process — a principle I continue to champion.

Why Figma is the most underrated AI tool for collaborative delivery

One of the most time-saving and underrated tools is Figma. I know I just mentioned it above, but it's a game changer — particularly its real-time collaborative whiteboarding and template integrations.

The impact is significant. It allows distributed teams to co-create, brainstorm, and iterate on ideas asynchronously or synchronously, which drastically reduces the back-and-forth that usually happens over email or long meetings. Teams can visualize workflows, map processes, and track decisions all in one place, making workshops and planning sessions much more efficient.

Another underrated aspect is its integration with Slack, which bridges the gap between AI in project planning and execution. This combination ensures that decisions and action items captured in collaborative sessions automatically feed into task tracking, reducing duplication of effort and accelerating follow-through.

The tool enables clarity, alignment, and faster decision-making across distributed teams, which is often underappreciated until you’ve run multiple collaborative projects across regions.

How AI acts as a copilot in modern delivery rituals

With AI increasingly becoming part of the delivery ecosystem, we’re rethinking traditional rituals to leverage AI’s strengths while keeping human judgment central.

• Aligning teams: Tools help surface misalignments or overlapping responsibilities, but alignment still relies on human discussion and AI in project decision making.
• Validating work: AI accelerates testing and simulation, flagging errors or compliance risks before manual reviews. For instance, in robotics and AI in project monitoring, predictive modeling can simulate edge-case scenarios that would take humans much longer to test.
• Managing execution: Routine tracking and reporting can now be automated, freeing time for the team to focus on problem-solving, strategy, and stakeholder engagement. Dashboards highlight risks or delays in real time, making stand-ups and retrospectives more focused on decisions rather than reporting.

The overarching shift is that delivery rituals are less about manual oversight and more about interpretation, judgment, and strategic guidance.

kateryna's tip

One of the most time-saving and underrated tools is Figma — particularly its real-time collaborative whiteboarding and template integrations.

How agentic workflows and orchestration platforms support complex delivery

When we experiment with orchestration platforms and agentic workflows, the focus is on areas that benefit most from coordination and predictive insight, such as:

• Task prioritization and dependency management: Using AI in project forecasting to identify which workstreams are critical, where bottlenecks might occur, and how resources should be allocated.
• Risk and compliance monitoring: Automating detection of potential compliance or safety issues in robotics and AI projects before they impact delivery.
• Cross-team communication: Streamlining updates and highlighting misalignments so teams spend less time chasing status updates.

As with AI in general, we've found that orchestration platforms still need a human in the loop for strategic decision-making and ethical considerations. The technology accelerates coordination and provides actionable insights, but it doesn’t replace the judgment required to balance priorities, navigate trade-offs, or engage stakeholders effectively.

Why ethics, inclusion, and diversity must shape AI-driven project delivery

Moving forward, I believe ethics, inclusion, and diversity will become core to how we define success.

AI and robotics can accelerate delivery and decision-making, but without intentional focus on ethical frameworks and diverse perspectives, projects risk reinforcing biases or creating unintended harm.

In practice, this means we must embed ethics and inclusivity into every stage of delivery — from defining scope, to designing systems, to iterating on feedback — and treating them as enablers of innovation rather than constraints.

Teams that integrate these principles early will not only build better products but also foster trust, adaptability, and long-term resilience.

How AI is moving teams away from rigid processes toward adaptive delivery

Project delivery itself will increasingly evolve from task tracking to strategic orchestration, with AI acting as a real-time copilot. Over the next five years, I expect AI to not just automate reporting or task updates, but to actively predict risks, recommend priorities, and even suggest trade-offs based on data from past projects, resource availability, and market signals.

Teams will shift away from rigid Gantt charts and heavy documentation, moving toward lightweight, adaptive workflows where humans focus on high-value activities: defining strategy, interpreting AI insights, making ethical decisions, and fostering collaboration. In essence, AI will handle the “what could happen” analysis, while product managers decide what should happen.

I also anticipate that cross-functional alignment and decision-making will accelerate, as AI can simulate outcomes of different delivery approaches in real time, allowing teams to iterate faster and innovate responsibly without compromising compliance or quality.

Ultimately, successful delivery will be defined by how effectively humans and AI collaborate, rather than how tightly we follow predefined processes.

kateryna's tip

My role is shifting from primarily “managing tasks” to orchestrating innovation responsibly and ensuring AI initiatives create real value without compromising safety, compliance, or trust.

How product leaders can adopt AI without losing human judgment

My advice to delivery leaders navigating this moment of change is to treat AI as a partner, not a replacement, while keeping ethics and diversity at the center of every decision. Use AI to automate repetitive tasks and surface insights, but ensure humans drive strategy, ethical considerations, and inclusive decision-making.

Also, foster diverse teams and perspectives — projects succeed when different experiences and viewpoints are considered, especially in AI and robotics where biases or blind spots can have real-world consequences.

And finally, invest in your team’s ability to work responsibly with AI, combining technical skill with ethical awareness and collaborative problem-solving. Leaders who balance innovation with inclusivity and integrity will be the ones who drive meaningful, sustainable impact.

Follow along

You can follow Kateryna Portmann's work on LinkedIn. And don't miss her initiative, Women in Robotics!

More expert interviews to come on The Digital Project Manager.