Practical Ways to Uplevel Project Delivery with AI, According to a Platform Engineering Lead
Adora Nwodo is a platform engineering lead, author, speaker, and the founder of NexaScale. Her focus is on using AI to enhance project delivery — primarily by reducing repetitive workflows and cognitive overload.
AI shifts delivery from administration to leadership: AI absorbs repetitive, high-volume work (summaries, triage, context gathering), freeing leaders to focus on architecture, alignment, mentoring, and decision quality.
The biggest wins come from layering AI onto existing tools: Productivity gains don’t require new platforms—adding AI to Jira, Slack, CI, and meetings dramatically reduces cognitive load and speeds up decisions.
AI creates predictable delivery: Six-week cycles, fewer rituals, reusable prompts, and AI-driven context replace heavy project management with clarity, momentum, and trust.
We sat down with Adora to learn how she's using AI, and how it's affecting her systems and rituals in practice. Here's what she had to say.
A career in building predictable project delivery
I am a platform engineering leader, public speaker, author of seven books, and founder of NexaScale — an initiative aimed at fostering the growth and development of technology enthusiasts. In my role today, I focus on building strong engineering foundations that help teams ship reliable products with speed and confidence.
I work closely with developers, SREs, and product teams to remove friction, improve delivery processes, and create the kind of platform experience that supports consistent and predictable project delivery across an organization.
Unlock for Free
Create a free account to finish this piece and join a community of forward-thinking leaders unlocking tools, playbooks, and insights for thriving in the age of AI.
Already a member? Log In
How AI is reshaping project delivery and engineering leadership
AI is changing my role in very real and practical ways. As a platform engineering leader, I now think about project delivery with two parallel tracks:
- One is about the traditional foundations like reliability, automation, and developer experience.
- The other is about how AI becomes part of the workflow itself.
I spend far less time manually reviewing repeatable work, digging through logs, or coordinating handoffs that can be automated. AI-driven tooling handles a lot of the first-pass work for things like documentation generation, test scaffolding, environment setup, and noisy operational alerts. I also spend less time answering the same operational questions because AI agents now provide guidance directly to engineers inside an internal platform.
What gets most of my attention now is shaping how teams use AI in a safe and strategic way. This includes choosing the right AI tools, designing responsible workflows, creating guardrails, and thinking about how AI changes engineering culture.
I also spend more time on higher-level system architecture, cross-team alignment, and the human side of delivery since AI clears space for deeper thinking:
- Higher-level system architecture: Instead of reacting to individual issues, I think in terms of long-term patterns and shared building blocks that make delivery faster and more resilient. To do that, I look holistically at how our services, workflows, and platforms scale as the business grows.
- Cross-team alignment: This shows up in a personal way for me. Since my own use of AI is ahead of where the broader team currently is, I spend more time translating what I learn from AI-assisted workflows into improvements in our platform, documentation, and internal tooling. This lets others benefit gradually without forcing change before they are ready.
- The human side of delivery: AI frees up space, so I invest more time in mentoring engineers, understanding their challenges, and creating an environment where they can focus on higher-impact work rather than dealing with repetitive tasks.
Overall, project delivery is becoming less about pushing tasks over the line and more about orchestrating an environment where intelligent automation, strong platforms, and skilled people work together with much less friction.
Project delivery is becoming less about pushing tasks over the line and more about orchestrating an environment where intelligent automation, strong platforms, and skilled people work together with much less friction.
How to identify the engineering workflows most ready for AI automation
The clearest opportunities for AI in project delivery right now sit in the high-volume work that drains time but doesn’t really require deep judgment. Things like log reading, ticket triage, documentation cleanup, dependency checks, and summarizing long technical threads are perfect candidates. These are areas where AI can be implemented directly through reusable prompts, automated agents in developer portals, or even simple integrations that run inside CI and surface insights before a human needs to step in.
AI also fits well into meeting prep and knowledge sharing. Tools like Gemini already help me turn long discussions into clear action points, and that same idea can be extended to sprint reviews, retros, and decision records. It gives teams better shared understanding without the usual overhead.
The parts that still need a human touch are the areas where context, trust, and intuition matter. Things like setting technical direction, evaluating trade-offs that affect people, making calls during ambiguous incidents, or coaching engineers through challenges. These moments depend on judgment and experience, and AI can only support — not replace — that.
So the split is simple. Let AI in project management handle the repetitive work that slows delivery down. Let humans handle the creative, strategic, and people-focused decisions that move delivery forward.
How AI creates a mindset shift
It's amazing how quickly AI in project task automation shifts from feeling like a nice extra to becoming part of the core workflow. I expected efficiency gains, but I didn’t expect it to change me.
I used to spend a lot of energy collecting context and trying to keep every detail in my head so I could lead with confidence. With AI supporting that work, I quickly noticed a mindset shift. I started showing up calmer and more prepared because I wasn't scrambling to remember everything. I could focus on direction, clarity, and the quality of decisions instead of the mechanics of gathering information.
This has been a really positive change. Work feels more intentional and less reactive. It also came naturally because I saw the benefits almost immediately. I did not plan for a mindset shift. It simply happened as I used AI to remove the noise and create more space for leadership.
How a lightweight six-week cycle improves delivery
I have been moving away from traditional project management approaches that rely on constant status meetings, rigid timelines, and heavy documentation. Instead, I use a more lightweight system built around clear cycles and high-quality context. The transition happened gradually — over time, I reduced the number of check-in meetings, shared shorter updates, and focused discussions on decisions rather than reporting.
One of the anchors of this system is our six-week roadmap review. Rather than trying to manage progress week by week in a detailed project plan, we align deeply every six weeks, agree on priorities, and set expectations with stakeholders. Between those cycles, the focus is on momentum, not paperwork.
Because we work with many stakeholders, the biggest shift has been how I manage the flow of information. Instead of building long status reports or sitting in multiple syncs, I use AI tools like Gemini to summarize Slack threads, meeting notes, documents, and updates from Jira. That gives me a clear picture of what is moving, what is blocked, and what decisions need attention without spending hours gathering context.
Of course, this means our rituals get lighter as well, allowing the team to focus on execution rather than constant reporting.
The result is a lighter process with more clarity. Stakeholders get better insight because the information is concise and timely, engineers have fewer interruptions, and I can spend my energy on direction and problem-solving instead of administrative tasks. The six-week cycle gives structure, and AI gives the day-to-day clarity that keeps everything moving.
How reusable engineering prompts reduce repetitive work
My team has started using reusable engineering prompts. We have created a small library of prompts that support common engineering tasks, such as writing unit tests, improving error messages, summarizing service behavior, or reviewing changes for potential reliability issues. These prompts are written once, tested, and then reused so engineers do not start from scratch every time they ask an assistant for help.
For example, one prompt is dedicated to generating structured release notes from pull requests. It expects specific inputs like the ticket number, area of the codebase, risk level, and customer impact. Another one focuses on translating operational alerts into plain language along with the most likely root causes based on previous incidents.
The setup was simple at first and lived in a shared document, and we are gradually moving the prompts into our internal platform so they become easier to discover and maintain. The effort to build and refine each prompt is small compared to the ongoing time saved, because the output is clearer, faster, and more consistent every time someone uses it.
It cuts down a lot of repetitive work and helps people move through delivery with more confidence.
How adding an AI layer to everyday engineering tools accelerates delivery
Right now, my delivery stack is pretty straightforward. I use Gemini, Jira, Slack, Google Workspace, and the usual engineering tools my team already works with — for example, GitHub for version control, our CI pipeline for builds and deploys, and Datadog for monitoring and alerting.
What has changed in the past year isn't the AI project management tools themselves; it's how I use them.
- Gemini has had the biggest impact. I use it for meeting prep, summarizing long documents, getting quick insights from logs, and turning notes into clear action items. It has replaced a lot of manual context gathering and helps me make decisions much faster.
- Jira remains the core of work tracking. I haven’t replaced it. What’s changed is that I manage my own workflow around it more efficiently because I can use Gemini to digest large epics, understand progress quickly, and prepare for planning sessions without spending hours reading through everything.
- Slack is still central to communication. But instead of trying to keep up with every thread, I lean on Slack's AI summaries to extract what’s important so I don’t get buried in messages. Slack's AI also answers questions and automates tasks within the platform.
So the tools themselves are largely the same, but my workflow has evolved. The biggest shift has been adding an AI layer on top of everyday tools. In other words, it's the way I pair Gemini with the tools I already use every day to interpret what those tools hold.
The biggest shift has been adding an AI layer on top of everyday tools. In other words, it’s the way I pair Gemini with the tools I already use every day to interpret what those tools hold.
How Gemini improves meetings, recall, and decision-making
Let's dive deeper into Gemini.
One of the most underrated capabilities right now is simply the ability to use AI to summarize meetings, long Slack threads, documents, and even what happened across the week. It sounds small, but it has been a huge time saver for managers.
It is not a flashy feature, but the impact is real. It gives back time, lowers stress, and keeps execution moving without delays caused by information overload. For managers, that has been one of the biggest wins.
In practice, I use Gemini heavily in my meetings — mainly as a support tool for meeting recall, rather than a system that thinks for me. I use it to prepare agendas, summarize discussions, track decisions, and pull insights from long threads or documents.
Most of my meetings are recorded, and Gemini automatically generates summaries so I don't have to rewatch long videos to collect the key points. When I review a summary, I refine it, clarify anything that was misunderstood, and extract the decisions and actions that actually matter.
I also use it to quickly search across multiple recordings if I need to confirm earlier context, locate the original reason for a decision, or pull out details from a discussion that happened weeks ago. It saves me from digging through hours of content, but I remain the one interpreting the information and driving the next steps.
All of this helps me walk into meetings with full context and leave with concise action points. And because I get clarity faster, I make decisions faster.
Practical experimentation with agentic workflows
We are experimenting with agentic workflows, but in a very practical way. Our main focus right now is the company-wide AI challenge, which is more of a hack-and-learn initiative than a formal platform effort. The entire company is building small AI projects to get hands-on experience and understand what agentic workflows could look like in our context.
Personally, for delivery work, I’m paying attention to the ideas involving agents that help with things like summarizing context, reducing repetitive tasks, or improving decision-making.
It’s still early, but it’s going well. People are learning fast, and we’re already seeing patterns that could scale into more structured agentic workflows later on.
Why AI will shift project management from administration to leadership
In the near future, project delivery will shift from being manager-driven to being context-driven. AI in project status reporting will handle most of the tracking, summarizing, and coordination work, which means teams will move based on real-time insight instead of scheduled check-ins. Managers will spend far less time running processes and far more time guiding strategy, coaching people, and making judgment calls.
The idea of a project manager as the primary source of updates will also fade. AI will surface progress, risks, and dependencies automatically. Teams will collaborate with AI agents the same way they collaborate with tools today. The result will be faster decisions, lighter rituals, and delivery systems that feel more adaptive than structured.
The human role won’t disappear, but it will shift toward leadership, clarity, and decision-making rather than administration.
Why AI won’t replace the way we work overnight
My advice is to start small, stay curious, and focus on clarity rather than control.
This moment is not about replacing the way we work overnight. It is about letting AI take on the repetitive parts of delivery so leaders can focus on people, direction, and decision-making.
The biggest unlock comes from experimenting with simple use cases like summaries, meeting prep, or context gathering. Those small wins build confidence and create space for deeper change.
And most importantly, keep your teams involved and keep the process light. AI works best when it supports the culture you already have, not when it becomes another layer of complexity.
Follow along
You can follow Adora Nwodo on LinkedIn, X, or her personal website. And check out NexaScale.
More expert interviews to come on The Digital Project Manager!
