Hello There Clippy

Welcome to my speaking page! When it comes to speaking publicly, I try to only talk about things I’m passionate about, and therefore know well, since it helps with eliminating anxiety.

Once you get rolling on something you love, you forget all the many things that might’ve made you scared about presenting in the first place, and the impostor syndrome rolls away. At least, sometimes it does.

My talks tend to stay in the realm of technology, but I’m always open to speaking on anything within my areas of interest. Scroll through to see what I’ve been up to. If you’d like me to speak at your event, get in touch!

Watch

Stuff from my YouTube channel of me talking about things.

Talks & Workshops

Here are some of the talks and workshops I’ve developed. I try to keep this section updated, but I’m always working on new ideas. In addition to scrolling this page, you can browse all of my published slide decks to get an idea of what I speak about.

Never Trust a Coding Agent That Only Reads the Code

Best For: Software engineers and anyone building with AI coding tools

Format: 45-60 minute talk (no laptops) or workshop (laptops required)

Much like life itself, building software is complicated. Any respectable harness (Claude Code, Codex, OpenCode, GitHub Copilot, or Antigravity) already indexes your codebase, but that’s only one part of a larger ecosystem. When your agent makes mistakes, it’s usually not because it sucks. It’s because it doesn’t have the full picture. Code shows how something works, not why. Your agent doesn’t automatically know what’s going on in your chats, backlog, cloud, databases, dashboards, or partner codebases.

To fix that, you’ve got to broaden its horizons. In this talk, we’ll walk through real-world scenarios where an agent with only codebase access falls short on everyday engineering challenges. You’ll see how to wire up MCP servers, path-scoped instructions, agent skills, and agent hooks to pair trusted facts with trusted evidence while governing read and write access responsibly.

Your agent can do so much more than edit code. With context engineering, it becomes a research, knowledge, and all-around engineering tool, helping you as the human in the loop. While you might not write it by hand anymore, your job is to deliver code you have proven to work, and that’s easier when your agent has the full story.

Much like life itself, building software is complicated. Any respectable harness (Claude Code, Codex, OpenCode, GitHub Copilot, or Antigravity) already indexes your codebase, but that’s only one part of a larger ecosystem. When your agent makes mistakes, it’s usually not because it sucks. It’s because it doesn’t have the full picture. Code shows how something works, not why. Your agent doesn’t automatically know what’s going on in your chats, backlog, cloud, databases, dashboards, or partner codebases.

To fix that, you’ve got to broaden its horizons. In this workshop, you’ll step into a simulated system where an agent with only codebase access falls short on everyday engineering challenges. Then we’ll wire up MCP servers, path-scoped instructions, agent skills, and agent hooks to pair trusted facts with trusted evidence while governing read and write access responsibly.

Your agent can do so much more than edit code. With context engineering, it becomes a research, knowledge, and all-around engineering tool, helping you as the human in the loop. While you might not write it by hand anymore, your job is to deliver code you have proven to work, and that’s easier when your agent has the full story.

Context Engineering Is Software Engineering

Best For: Software engineers and anyone building with AI coding tools

Format: 45-60 minute talk

Codex or Claude Code. Cursor or GitHub Copilot. Terminal or IDE. If you’re wondering which of these gives you the best outputs in AI-assisted coding, the answer is all of them, or none of them. Maybe you fancy Google’s Antigravity instead, or the desktop app versions of any of the above. It hardly matters which you use, because your coding agent is only as good as the context you engineer around it.

In this talk we’ll define what context engineering is, how it works, what makes it portable across harnesses and models, and why it’s as important as the code itself. The cool kids are all saying context is the new code, and they’re right, which is why it deserves the same rigor we’ve always given the code. We’ll cover terms like prompt, harness, loop, and graph engineering, and why your context needs tests, just like your code.

By the end, you’ll understand your options for durable context, including custom instructions, custom agents, agent skills, agent hooks, agent plugins, and MCP servers. And you’ll see why context engineering, at its core, is software engineering, with a desperate need for humans in the loop. You might not need to read the code anymore, but you really do need to read the context.

The Children Yearn For the Mines: Finding Craftsmanship in AI-Assisted Software Engineering

Best For: Software engineers and anyone building with AI coding tools

Format: 30-45 minute talk

For many of us software engineers, writing code was how we understood a system, shaped it, and found joy in our work. As AI writes more of the code, some of us feel more creative than ever, while others feel we’ve lost the very thing we entered the profession to do.

The “children yearn for the mines” meme jokes that child labor laws got kids out of dangerous physical mines, only for them to happily pick up virtual pickaxes in Minecraft. The game can auto-generate a world for them, but the fun is in building it themselves and expressing their creativity. So when AI generates the code, how do we keep that same connection to the systems we’re creating?

In this talk, we’ll explore ways to maintain cognitive ownership of systems built with agents, using examples like Simon Willison’s Agentic Engineering Patterns, Addy Osmani’s agent skills, and GitHub’s Awesome Copilot, where developers build and test context like code. We’ll also use Andrej Karpathy’s distinction between loving coding and building to examine why some enjoy this shift while others feel reduced to “looks good to me” merchants. You’ll leave with resources for understanding AI-generated code and reassurance that you’re not alone in seeking a way back to craftsmanship. A way back to the mines.

Agentic AI: From Acronyms to Applications

Best For: Technical and non-technical audiences, from college students to working professionals

Format: 45-60 minute talk (no laptops) or workshop (laptops required)

With the rise of Agentic AI, Uncle Ben’s “with great power comes great responsibility” warning has never felt more relevant. AI agents can gather information, use tools, make decisions, and take action with limited human direction. Everybody’s talking about them, but ask 10 people what an agent is and you’ll get 10 different answers. So what actually makes a system agentic, how is it different from Generative AI, and where does human responsibility still lie?

In this session, we’ll cut through the hype by looking at how AI agents are being used in the real world, and what happens when they go wrong. Through interactive scenarios where an agent is given a goal, a prompt, and access to tools, we’ll unpack why a seemingly good result still went sideways. Along the way, we’ll talk candidly about AI costs, layoffs, hiring trends, and the future of work.

By the end, you’ll understand LLMs, RAG, MCP, RLHF, context, and context windows through practical examples. You’ll see what Agentic AI can do, where it falls short, and why the growing field of AI Engineering still needs humans in the loop.

With the rise of Agentic AI, Uncle Ben’s “with great power comes great responsibility” warning has never felt more relevant. AI agents can gather information, use tools, make decisions, and take action with limited human direction. Everybody’s talking about them, but ask 10 people what an agent is and you’ll get 10 different answers. So what makes a system agentic, how is it different from Generative AI, and where does human responsibility still lie?

In this workshop, we’ll cut through the hype with an approachable look at how modern AI works, then jump into a guided lab. You’ll start with a bare AI model to see how little it can do on its own, and connect tools and context so it can grab live info and get things done. Next, you’ll lock down a risky agent with guardrails and human approval, before building your own custom agent. Along the way, we’ll talk candidly about AI costs, layoffs, hiring trends, and the future of work.

By the end, you’ll understand terms like LLMs, RAG, MCP, RLHF, context, and context windows firsthand. You’ll see what Agentic AI can do, where it falls short, how agents work across the leading platforms, and why AI still needs humans in the loop.

Presented At: SEO Career Summit 2026 · SEO TechDevCon 2026 · UNC Charlotte Women in Data Science Conference 2026 · PMI Carolinas Professional Development Day 2025 · Numerous Discovery Days at Microsoft Charlotte

Resources: Companion Article · Hands On Workshop

How to Train Your AI: Demystifying ChatGPT With Machine Learning Basics

Best For: Technical and non-technical audiences, from high school students to working professionals

Format: 45-60 minute talk (no laptops) or workshop (laptops required)

ChatGPT can feel like magic, but underneath it all, it’s making highly educated guesses based on patterns. This AI Literacy talk takes an intentionally oversimplified dive into machine learning, neural networks, and deep learning by turning you into the computer.

You’ll write traditional programming rules for recognizing a cat, learn to recognize an unfamiliar writing system from examples, and guess missing words in sentences. Along the way, you’ll get an approachable understanding of supervised learning, self-supervised learning, and how large language models (LLMs) learn by guessing, checking, and adjusting across enormous amounts of text.

By the end, you’ll understand how examples become patterns, how patterns become predictions, and what makes training AI models so expensive. The result may seem intelligent, but it’s fundamentally pattern recognition at scale and still requires humans in the loop.

ChatGPT can feel like magic, but underneath it all, it’s making highly educated guesses based on patterns. This AI Literacy workshop takes an intentionally oversimplified dive into machine learning, neural networks, and deep learning by turning you into the computer.

You’ll write traditional programming rules for recognizing a cat and guess missing words in sentences. Then, in a visual, browser-based exercise, you’ll tune the inputs and parameters to a model and see how they affect the output. Along the way, you’ll get an approachable understanding of supervised learning, self-supervised learning, and how large language models (LLMs) learn by guessing, checking, and adjusting across enormous amounts of text.

By the end, you’ll understand how examples become patterns, how patterns become predictions, and what makes training AI models so expensive. The result may seem intelligent, but it’s fundamentally pattern recognition at scale and still requires humans in the loop.

Presented At: Numerous Discovery Days at Microsoft Charlotte, including this one with Carolina Youth Coalition · PMI Carolinas 2025 Kickoff · Demystifying AI With Techsgiving

Resources: Companion Article · Facilitation Guide · Inspired by Parand Tony Dugar’s articles on AI, ChatGPT, and Deep Learning

AI Technical Interview Workshop

Best For: College students studying computer science or related majors

Format: 60 minute workshop (laptops required)

AI-assisted software development is everywhere now. Companies like Canva, Meta, and Google let candidates use AI in coding interviews, and teams at Anthropic are dropping data structures and algorithms questions altogether. So if AI writes the code, what are interviewers actually evaluating, and how do you prove you can do the job?

In this workshop, you’ll step into a realistic AI-assisted interview environment where an AI agent has already written the code. Across a series of challenges, you’re given code that compiles and looks right, but has a logical disconnect between the problem requirements and what the agent actually built. You can use a built-in AI assistant to ask questions and make changes, with every prompt and response visible to the interviewer.

To succeed, you have to read the requirements, understand what the code is doing, spot where the agent’s logic went wrong, and guide the AI to fix it. AI can write code, but it can’t do your job. Your job is to deliver code you have proven to work.

Presented At: UNC Charlotte Industry Day · Microsoft Charlotte AI Engineering Summit With Central Piedmont Community College

Resources: Workshop Website · How This Workshop Was Built · GitHub Repository · Companion Article

We're Not Cooked: Your Tech Career Survival Guide in the Age of AI

Best For: College students studying computer science or related majors

Format: 45-60 minute talk

There’s never been a scarier time to be a computer science student, with tech CEOs fearmongering that AI will replace all white-collar jobs or that software engineers will be replaced in months, even as their own companies aggressively hire them and walk back their job apocalypse prophecies. So how do you get a tech job now that AI can write code, and is a computer science degree still worth it? I mean…chat, are we cooked? Not at all!

In this talk, we’ll start with a quick history of AI and a live next-token prediction exercise to see how LLMs work under the hood. From there, we’ll examine why writing code was never the real bottleneck in software engineering, and follow the money through GPUs, data centers, and layoffs. We’ll cut through AI washing to see why companies are cutting costs to pay for ridiculously expensive compute, and why some employers are already rehiring the workers they thought AI could replace.

You’ll also meet AI Engineering, the fastest-growing job on LinkedIn, and see why it’s much more accessible than Machine Learning. We’ll cover the AI terms you should know, which technical skills and certifications still matter, how AI resume screeners and interviews work, and what a resume that landed a Microsoft software engineering internship looks like. Coding manually is dead, but the judgment, systems thinking, and accountability of software engineering are becoming more valuable, not less. Times might be changing, but we aren’t cooked. Far from it.

Presented At: Discovery Day with UNC Charlotte · Discovery Day with NC A&T and Winston-Salem State University · SEO TechDevCon 2025

Resources: Companion Article · AI & Your Tech Career · The Rise of the AI Engineer · The Concerns That Inspired the Talk

Precision Meets Creativity with Microsoft Designer

Best For: Middle and high school students of all backgrounds

Format: 60 minute workshop (laptops required)

AI can generate an image in seconds, but creativity still belongs to the human behind the prompt. In this workshop, you’ll start by matching an AI-generated image to two possible prompts, one vague and one detailed, to see how specificity, clarity, and creativity shape the result. Then you’ll use structured, fill-in-the-blank prompts to create your own visuals with Microsoft Designer, leaving with something you conceived and directed, and a clear understanding that AI supports human creativity rather than replacing it.

Presented At: Numerous Discovery Days at Microsoft Charlotte · Charlotte-Mecklenburg Schools Bases Loaded 2025 · Charlotte-Mecklenburg Schools Bases Loaded 2026 · Black at Microsoft Minority Student Day · NSBE Charlotte STEM Fair

Resources: Companion Article · Workshop Website

Context Is All You Need: AI Engineering with the Petoi 'Bittle X' Robot Dog

Best For: High school and college students interested in AI, programming, or robotics

Format: 60 minute workshop (laptops required)

AI models can be brilliant, but they need specific context to be useful. In this workshop, you’ll start with an AI chat interface that only knows how to send three basic commands to a Petoi ‘Bittle X’ Robot Dog. Your job as an AI Engineer is to teach it what else the robot can do.

Using Python, VS Code, and GitHub Copilot, you’ll read technical documentation, expose what the AI sends to the hardware, chain simple actions into complex behaviors, and connect the model to live documentation through a simplified version of retrieval-augmented generation (RAG).

By the end, you’ll have seen both sides of AI Engineering. Traditional code controls the robot directly, while context teaches AI how to control it.

Presented At: Charlotte-Mecklenburg Schools Bases Loaded 2025 · Carolina Youth Coalition Discovery Day with recaps from Microsoft and Petoi

Resources: Companion Article · Workshop Code · Facilitation Guide · Petoi Robotics Curriculum Feature · Petoi Workshop Feature

Talk Details

Features, Mentions, & Testimonials

Stuff from around the internet where I’ve been featured, mentioned, or talked about. Just keeping track of other people talking about me so I don’t seem too self-aggrandizing on this page.

Segun Akinyemi captivated the entire room, memes and all. His presentation on Agentic AI and the evolving landscape of AI engineering roles was one of the most informative sessions I’ve been to, and left me genuinely hopeful about where tech careers are headed.

SEO TechDevCon 2026 Attendee

View Source
Segun Akinyemi speaking with attendees at the Charlotte Business Journal 40 Under 40 Awards

Charlotte Business Journal

40 Under 40: Segun Akinyemi, Microsoft Corporation

Charlotte Business Journal’s 2025 40 Under 40 profile of Segun Akinyemi.

More Items

26 items

Segun Akinyemi delivered an engaging, interactive presentation that challenged some of my assumptions about AI, particularly its environmental impact and its role in the future of work. I was especially interested in learning how AI can advance existing industries and create new possibilities rather than simply replace the roles within them.

Attendee of Microsoft Charlotte Discovery Day with UNC Charlotte October 2026

View Source

One of my favorite moments from the summit happened after the final session, when a few peers and I had the chance to continue a conversation with Mr. Akinyemi outside the event. What started as a casual conversation turned into a meaningful discussion about technology, AI, and career growth. Walking away from that conversation, I realized how powerful it is to connect with people beyond the stage and simply be curious. It was such a genuine interaction, and I learned so much from it.

SEO Career Summit 2026 Attendee

View Source

Beyond that I was surprised by the valuable insights I gained from all the panelists during, a big one from Segun Akinyemi being that I won’t be out of a job, discovering that Agentic AI is just another tool in my arsenal as an aspiring Software Engineer that encapsulates all these new AI-related “roles” (like Prompt Engineering).

SEO TechDevCon 2026 Attendee

View Source
Power Automate surrounded by connected automation icons

Microsoft Blog

Simplifying Nonprofit Volunteering at Microsoft with Power Automate

Segun Akinyemi shares how he uses Microsoft Power Platform and Power Automate to help manage his work-life-volunteering balance.

Segun Akinyemi - Thank you for reminding me that tech is a marathon of learning new tools, where a strong foundation keeps you running. Plus, learning public speaking is something I’m excited to take back and refine in my student org and leadership.

SEO TechDevCon 2026 Attendee

View Source

Shoutout to my favorite panelist Segun Akinyemi (although I didn’t get any Microsoft merch) whose humor, insights, and energy about AI and its misconceptions put to rest any doubt inside me to pursue a career in tech. This has made me ponder where I would want to add value in my career.

SEO TechDevCon 2026 Attendee

View Source
Two Petoi Bittle robot dogs under pink and blue lighting

Petoi

Bittle Python Robotics Curriculum & Microsoft Event Focusing On K12 STEM Education

Petoi features Segun Akinyemi’s three-level programming curriculum and Microsoft Discovery Day workshop using Bittle X robot dogs.

WCNC Charlotte

Microsoft Hosts Students for Discovery Day

Bailey Middle School students visited Microsoft Charlotte for Discovery Day, a field trip program founded in 2022 by Microsoft software engineer Segun Akinyemi to connect local students with tech careers and hands on learning.

The hands-on session on Agentic AI by Segun Akinyemi was particularly insightful. It was interesting to see how agent-based systems are orchestrated in practice, from real-time decision-making and API interactions to configuring tool usage and instruction layers for more controlled, context-aware behavior.

UNC Charlotte Women in Data Science Conference 2026 Attendee

View Source
LinkedIn post

A recap of Black at Microsoft’s Minority Student Day, a STEM education event I led for students from high schools across Charlotte.

Media & Bio

All this stuff is mine and available for use. Just let me know if you use anything!

Headshot

Segun Akinyemi
Segun Akinyemi Download

Short Bio

Segun Akinyemi is a Nigerian American writer, speaker, educator, and Senior Software Engineer at Microsoft known for explaining complex topics in a practical, down-to-earth way. He writes, speaks, and teaches about software engineering, applied AI, and tech careers, helping students and professionals use AI tools while keeping humans in the loop. Based in Charlotte, he founded Discovery Days, a local field trip program connecting students with STEM opportunities, created and maintains Charlotte Third Places, and was named to the 2025 Charlotte Business Journal 40 Under 40 list at age 28. Learn more at segunak.com.

Long Bio

Segun Akinyemi is a Nigerian American writer, speaker, educator, and Senior Software Engineer at Microsoft known for explaining complex topics in a practical, down-to-earth way. Born in Seattle, educated in St. Louis, and now based in Charlotte, he connects people with knowledge, community, and opportunity through technology.

At heart, he loves education and creating approachable learning experiences, deepening his own understanding along the way. Grounded in his Christian faith, he founded Discovery Days, a field trip program connecting local students with STEM opportunities, and built Charlotte Third Places, a guide to welcoming community spaces around the city. These efforts contributed to his recognition on the 2025 Charlotte Business Journal 40 Under 40 list at age 28.

At Microsoft and beyond, he writes, speaks, and teaches about software engineering, applied AI, and tech careers, helping students and professionals use AI tools while keeping humans in the loop. Outside of work, some of his favorite pursuits include reading, blogging, spending time at third places, searching for the best cinnamon rolls, exploring science fiction and fantasy, and watching anime (Steins;Gate remains his all time favorite). Learn more at segunak.com.

Long Bio

Get In Touch

Email address hidden

You can also reach me through my contact page or find me on Sessionize. For a more detailed record of what I have been up to, speaking and otherwise, browse my highlights.