Inspire AI: Transforming RVA Through Technology and Automation

AI Ready RVA
  • 5.0 (1)
  • TECHNOLOGY

Our mission is to cultivate AI literacy in the Greater Richmond Region through awareness, community engagement, education, and advocacy. In this podcast, we spotlight companies and individuals in the region who are pioneering the development and use of AI. 

  1. 3D AGO

    Ep 70 - The Rise Of AI Runtimes: Google's Agent Development Kit

    Send a text AI is starting to feel less like a feature you bolt onto a product and more like a system you have to run. That shift is easy to miss until you try to build something real: a workflow that calls APIs, keeps context across sessions, coordinates tasks, pauses for human approval, and resumes later without breaking. Suddenly prompts are not the hard part. Architecture is. I walk through what Google’s Agent Development Kit (ADK) reveals about the future of AI agents and agentic workflows. The core idea is event driven execution: a runner orchestrates the system while an agent emits events like “use this tool,” “update state,” “store an artifact,” or “request confirmation.” It’s a clean mental model for building an AI runtime with resumable execution, observable state, and tool integration that can actually survive production. We also get practical about agent design. Not every agent should be an LLM free styling its way through a task. I break down LLM agents for reasoning, workflow agents for deterministic reliability, and custom agents for complex orchestration, then connect that to the deeper takeaway: the model is the decision engine, but tools are the capability. Rich tool ecosystems and clear interfaces will matter more than chasing ever larger parameter counts. Finally, we talk governance and safety. Tool confirmation and human in the loop controls are not optional if agents can send emails, change data, or trigger real world actions. If you’re a leader, builder, or architect trying to scale enterprise AI responsibly, this is the mindset shift to make now. Subscribe, share this with a teammate, and leave a review with the guardrail you think every AI agent should have. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    11 min

  2. MAR 9

    Ep 69 - Future of Work: When Intelligence Lives In Your Tools

    Send a text What if the real shift in the future of work isn’t learning to code, but learning to supervise? We dig into a new operating model where product and engineering leaders step into the execution loop by directing AI coding agents that read repos, edit files, run tests, and open pull requests—while engineers safeguard architecture and correctness. The payoff is leverage: clear intent, tighter feedback loops, and artifacts that move from concept to code without the slow drag of endless handoffs. We break down the workflows that change first. Technical discovery goes from week‑long spelunking to safe, read‑only scans that map modules, APIs, logs, and risks. Strategy stops living in slides as agents draft API contracts, edge cases, rollout plans, observability requirements, and acceptance tests tailored to your repo conventions. Prototyping accelerates with feature‑flagged walking skeletons that ship telemetry and a passing test, so feasibility debates turn into concrete PR reviews. Communication gets sharper as release notes and risk flags are generated from diffs, not guesswork. Verification becomes culture when prompts encode done as tests pass with outputs shown, and CI automations become structured, maintainable flows rather than fragile hacks. Even roadmap hygiene matures as agents link traceability, standardize acceptance criteria, and rewrite unclear tasks. Speed without rigor is a trap, so we name the metrics that actually show progress: cycle time, change failure rate, experiment throughput, avoided defects, and review latency. We also surface the new risk surface—hallucinations and silent failures, security and supply chain exposure, data retention and IP policy mismatch, skill and ownership drift—and share pragmatic governance: permission scopes, sandboxing, allow‑listed integrations, audit logs, and mandatory human PR review. Tools like Claude Code, Codex, Cursor, and Windsurf are signals of a broader pattern: intelligence becoming ambient inside production systems. The winners won’t be the teams that chase the latest tool; they’ll be the ones who redesign workflows thoughtfully, measurably, and ethically. Join us as we turn leadership judgment into the core advantage: delegating to agents, specifying constraints and verification, and building execution loops that turn clarity into shipping code. If this resonates, follow the show, share it with a teammate who owns delivery, and leave a quick review telling us which workflow you want us to demo next. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    10 min

  3. MAR 2

    Ep 68 - Intent Over Keystrokes: The New Mind of the Modern Builder w/ Godwin Josh

    Send a text What happens when coding stops being about keystrokes and starts being about intent? We sit down with Godwin Josh—mentor, builder, and author of The New Mind—to unpack how agentic AI is transforming the path from student to work-ready engineer. Instead of celebrating speed for its own sake, we look at why tools like Windsurf, Claude Code, and Copilot accelerate learning, make patterns visible, and free developers to focus on judgment, problem framing, and real outcomes. We trace Godwin’s journey from early DOS animations and hardware products to AI-first teams, then dive into a practical stack for modern builders: Linux for environments, Python for ecosystem depth, and an agentic layer that includes skills, agents.md for self-describing projects, and soul.md for consistent behavior around testing, security, and clarity. With MCP acting like a universal “USB port,” models can discover and use tools reliably, turning agents into capable collaborators rather than autocomplete toys. The shift is profound: a developer becomes a director—defining goals, curating capabilities, and validating results—while agents handle scaffolding, refactors, and repetitive glue work. Mentorship emerges as the quiet engine behind impact. Raw intelligence doesn’t guarantee results; exposure to constraints, wise counsel, and clear goals does. We talk about building cross-disciplinary teams with universities, where physics meets data science and bio meets compute, and how AI compresses the learning curve so students can build real systems before graduation. We also confront the anxiety many veterans feel: when the “how” is automated, your edge becomes asking sharper questions, making faster decisions, and communicating with courage. Math and language prove durable; specific tools churn. If you’re a student, lead, or educator navigating agentic AI, you’ll leave with a playbook: codify standards in skills, describe projects with agents.md, shape agent behavior with soul.md, validate across multiple models, and measure progress by shipped value. Subscribe, share this with someone who’s rethinking their workflow, and leave a review telling us: what skill matters most when AI writes the boilerplate? Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    48 min

  4. FEB 23

    Ep 67 - RAG Done Right: Measure The Evidence Or Drift Into Error

    Send a text What happens when a brilliant-sounding AI gives the wrong answer with total confidence? We dig into the quiet culprit behind so many “LLM failures”: retrieval. Rather than judging how smart a model sounds, we walk through how to judge whether it looked at the right evidence, why that matters in high-stakes domains like finance, healthcare, HR, and government, and how leaders can stop organizational drift driven by outdated or partial sources. We break down four pillars every RAG team should track: retrieval precision and recall to balance noise versus coverage; context relevance and coverage to ensure the retrieved passages actually answer the question; groundedness and fluency so every claim traces back to evidence; and accuracy and completeness to catch stale or missing knowledge. Along the way, we share real-world patterns—chatbots citing old HR policies, assistants using superseded regulations, and tools surfacing obsolete medical guidance—and show how these errors spread when confidence outruns curation. Then we get practical. We outline precision@K and recall@K, golden question sets tied to authoritative documents, LLM-based judging for relevance and groundedness, and continuous regression testing as knowledge bases evolve. More importantly, we frame the cultural shift: assign ownership for knowledge freshness, make sources visible next to answers, and normalize verification at every level. Treat AI answers as drafts, retrieval as evidence, and evaluation as the safeguard. If you’re running or planning a RAG system, start by asking to see retrieved sources, build a small high-stakes golden set, and set a cadence for archiving and updates. If this conversation helped sharpen your approach to reliable AI, subscribe, share with a teammate who manages content or compliance, and leave a quick review with one insight you’re taking back to your team. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    10 min

  5. FEB 16

    Ep 66 - From Prompts To Process: Building Trustworthy AI Workflows w/ Tianzhen Lin

    Send a text When intelligence is everywhere but correctness is scarce, how do we lead without cutting corners? We sit down with Tianzhen (Tangent) Lin—veteran engineer and systems thinker—to unpack a practical, durable approach to building AI‑assisted products that hold up under pressure. No hype, no shortcuts: just the patterns that make teams faster and safer at the same time. We start by reframing large language models as “eager interns”: fast, helpful, and prone to saying yes. That mental model shifts responsibility back where it belongs—on leaders who must design workflows that surface assumptions, constrain degrees of freedom, and verify outcomes. Tangent explains why context remains a finite resource even with giant windows and how the “lost in the middle” effect undermines long prompts. The fix isn’t more chat; it’s better scaffolding. Specs, plans, and documentation become the backbone for repeatable success because they compress what matters and travel across sessions and teammates. From there, we dig into decomposition as a risk strategy. Breaking work into small, testable steps gives you early checkpoints to catch hallucinated requirements, unsafe libraries, or performance traps—like UI freezes from naive million‑row operations. Tangent shares a late‑night pivot where a strong, technology‑agnostic spec let the team re‑architect in hours, not days, turning a potential rewrite into a near‑seamless transition. We dive into verification as a non‑negotiable, the value of documentation as compressed context, and how institutional knowledge prevents the “sandcastle effect” when requirements shift or the tide comes in. The result is a playbook for leaders navigating an AI‑accelerated world: treat context like budget, invest in durable artifacts, decompose to control risk, and verify relentlessly. Do that well and AI stops being a confident amateur and starts acting like a reliable teammate. If you’re serious about trust, safety, and scalable speed, this conversation will sharpen your judgment and strengthen your systems. Subscribe, share with a teammate who ships software, and leave a review with the one workflow change you’ll make this week. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    33 min

  6. FEB 9

    Ep 65 - LLM-as-a-Judge: Evaluations That Scale

    Send a text What if your AI had a never-tired reviewer that caught quiet errors before they reached customers? We dive into LLM-as-judge—the simple but powerful pattern where one model generates and another evaluates—to show how leaders can scale quality without surrendering standards. From summaries that must capture the one sentence that matters to support answers that need to be grounded, safe, and on-brand, we break down where this approach shines and where it can fail you. We get practical with three evaluation formats—single-answer grading, pairwise comparisons, and reference-guided checks—and explain why ranking often beats raw scoring for stability. Then we map the biggest failure modes: confident nonsense that looks authoritative, biases you never asked for, and the danger of outsourcing values to a model’s defaults. The fix is leadership: define what good means, encode it in a rubric with clear anchors, and validate against human judgment before trusting the system. You’ll hear step-by-step patterns you can run next week: build a rubric with accuracy, groundedness, clarity, tone, safety, and actionability; use pairwise comparisons for model or draft selection; enable “jury mode” by aggregating multiple judgments; and force citations to specific source passages for verification over vibes. We also show how specialized judges—for factuality, tone, and compliance—reduce noise and improve reliability, and how monitoring helps you detect drift, compare model upgrades, and standardize quality across teams. If you’re ready to move from “we sometimes use AI” to “we operate AI inside a quality system,” this conversation gives you the mental models and playbooks to start. Subscribe, share with a teammate who ships AI features, and leave a review with one value you’d encode in your rubric. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    15 min

  7. FEB 2

    Ep 64 - Intelligence, Accountability, And You: From AI Slop to Sound Judgement

    Send a text The pace of AI can feel exhilarating until a polished report collapses under scrutiny and your team spends hours repairing “work slop.” We’re seeing a quiet shift across organizations: as intelligence becomes ambient, leadership’s edge moves from gathering information to evaluating it. That shift changes how we make calls, how we manage risk, and how we design trust into everyday workflows. We unpack practical decision hygiene that keeps speed from steamrolling substance. Treat AI outputs as drafts, not verdicts; verify facts, pressure-test conclusions, and define what “done” really means so polish doesn’t masquerade as insight. We share question prompts to expose missing data and faulty assumptions, and we draw clear lines between decision support and decision replacement—because confidence is not correctness, and accountability cannot be delegated to an algorithm. We then move into risk management where leaders operate as the safety net between model outputs and real-world consequences. From finance to healthcare to marketing, we outline why high-stakes decisions demand human in the loop and how to establish reviews, stress tests, and override paths without smothering speed. You don’t need to build models to lead well; you need to know where they break, how bias creeps in, and which failure modes matter for money, health, fairness, and reputation. Finally, we design for trust. Adoption accelerates when people know where AI is used, who stays accountable, and how decisions align with values. We explore transparency, explainability, and psychological safety so teams feel augmented rather than quietly judged or replaced. The throughline is simple: AI can generate options, but it can’t weigh meaning or carry consequence. That’s your job. If you’re ready to turn ambient intelligence into durable advantage, join us and upgrade your role to evaluator in chief. Enjoy the conversation? Follow the show, share with a colleague, and leave a quick review—then tell us the one change you’ll make to improve AI evaluation on your team. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    14 min

  8. JAN 26

    Ep 63 - Human In The Loop: Designing The Boundary Between Machines And Humans

    Send a text The moment an AI agent can issue refunds or change accounts, the conversation shifts from capability to responsibility. We dig into how to design trust between people and machines by choosing the right oversight model for the job: human in the loop for high-stakes decisions and human on the loop for fast, high-volume work. Along the way, we unpack concrete playbooks for customer service leaders and operators who need speed without sacrificing judgment. We start by drawing a clear line between decision-time approval and supervisory control, then show how confidence-based escalation creates dynamic autonomy. Instead of all-or-nothing automation, we use signals like model confidence, customer sentiment, value at risk, and ambiguity to route actions for auto-resolution or human review. We also break down synchronous versus asynchronous oversight, and why advanced teams separate planning (human approved) from execution (AI driven) to combine safety with scale. The examples ground the theory: a retailer that automated 40 percent of inquiries while escalating emotionally charged cases, an airline that trained its system through human corrections before handing off routine tickets, and insurers that pay clean claims instantly while auditing edge cases. You’ll hear a pragmatic checklist for safe scaling: map risk before tasks, set thresholds, give reviewers explanations, log everything, prevent automation bias, and train people to be AI supervisors. The goal isn’t to remove humans; it’s to elevate them—letting AI handle speed and repetition while humans guard empathy, accountability, and trust. Ready to build AI that knows when to ask for help? Subscribe, share this episode with a teammate, and leave a review with your top escalation trigger—we’ll feature the best ideas in a future show. Want to join a community of AI learners and enthusiasts? AI Ready RVA is leading the conversation and is rapidly rising as a hub for AI in the Richmond Region. Become a member and support our AI literacy initiatives.

    11 min
See All (72)

About

Our mission is to cultivate AI literacy in the Greater Richmond Region through awareness, community engagement, education, and advocacy. In this podcast, we spotlight companies and individuals in the region who are pioneering the development and use of AI. 

Information