Agents of Dev - AI, Agent & Agentic Development

The Futurum Group

The Agents of Dev Podcast explores how software developers build and will build applications in the era of AI, agents, and agentic technologies. Futurum Research analysts Mitch Ashley and Brad Shimmin bring their dual perspectives as industry analysts and practitioners to unpack what industry moves, vendor strategies, and on-the-ground engineering realities signal about the future of enterprise software. Each episode explores innovations, opportunities, and even oopsies to inform us about how AI-native development is reshaping the work of planning, designing, building, deploying, and operating software. Part of the Futurum Group family of podcasts.

Play

Open in

26 episodes
weekly
Avg 45 min
English

Episodes 26

Newest Oldest Longest

All20 Unplayed20 In progress0 Played0

Counted on this page — what you have heard stays on this device, so it is not something the list can be paged by.

#38
September 17 · 49 min
Beyond the AI Doomer Hype Harness Engineering, Context Economics, and Agentic Workflows | Agents of Dev Episode 38
Is AI heading toward a dystopian sci-fi nightmare, or are we just watching market hype mask the real architectural shifts happening in software engineering? In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin move past the polarities of AI doomerism vs. pure hype to tackle the pragmatic reality of agentic workflows. They analyze Cloudera’s enterprise sovereignty partnership with Mistral, examine Atlassian’s push into agentic SDLC management, and take a deep dive into the world of developer harnesses—comparing setups like Pi, Aider, and Maki. Along the way, they unpack context tokenomics, cash ownership, and why managing LLM memory often feels like living out the movie Memento. This and more on Agents of Dev, part of The Futurum Podcast Network.
#37
September 10 · 52 min
AWS DuckDB Moves, Fixing "Lazy" AI Agents, and the 2026 AI Budget Reckoning | Agents of Dev Episode 37
Email your questions and topics to the podcast here. Are enterprise AI budgets running out of control while agentic models pretend to finish tasks they haven't actually completed? In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin tackle the biggest developments across cloud infrastructure, AI behavior engineering, and enterprise data strategy. They start with AWS’s acquisition of Duck Labs and what embedding DuckDB directly into storage tiers means for the open-source ecosystem. Next, they unpack the issue of AI "phantom completeness" and how new tools like the Unlazy extension enforce behavioral discipline on agents. The pair also analyze Brad’s newly published Data Intelligence Platforms Signal report, break down Bill Gates’ 6,000-word call for AI guardrails and job protections, and discuss the "fund AI anyway" spending habits shaping IT budgets in 2026. This and more on Agents of Dev, part of The Futurum Podcast Network.
#36
September 3 · 49 min
Inside the Mind of an Open Source Innovator with Tracy Ragan | Agents of Dev Episode 36
Unvetted AI code generation in your software pipeline is like handing a construction apprentice the keys to a wrecking crane—it might get the job done, but it could just as easily knock down the entire building. In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin sit down with open source pioneer Tracy Ragan (CEO of DeployHub and Ortilius community leader) to dissect the state of the modern software supply chain. Tracy breaks down why hardcoded scripts continue to paralyze DevOps evolution, how global vulnerability collection laws are reshaping cybersecurity risk, and why "reachability" matters more than static SBOM scans. Together, the team explores the inevitable shift from bloated frontier models to lean, domain-specific small language models, proving why human collaboration still forms the ultimate zero-trust defense. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#35
August 27 · 48 min
Model Fatigue, Phantom Fixes, & Toolchain Chaos: Finding Control in Agentic Dev | Agents of Dev Episode 35
Are constantly shifting AI models and phantom fixes driving you to throw away hours of work? In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin dive into the real-world frustrations of developer toolchain chaos, model instability, and aggressive pricing shifts in the race for AI dominance. Brad highlights Oracle’s new APEX AI Application Generator and APEX Lang framework, while Mitch breaks down Dynatrace’s acquisition of Arise and why observability must anchor itself at the intent stage. Together, they demystify what actually defines a developer sandbox, test-drive Unsloth’s new agentic desktop harness, and examine Atlassian’s Code Context updates—offering a much-needed dose of tech therapy for software builders navigating today's rapid model churn. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#34
August 20 · 46 min
DynamoDB Vectors, Shift Zero Security, and Taming AI Tech Debt | Agents of Dev Episode 34
Email your questions and topics to the podcast here. As AI moves deeper into software development, keeping up with rapid updates can feel like chasing a moving target. In this episode of Agents of Dev, hosts Mitch Ashley and Brad Shimmin break down the biggest shifts transforming the developer ecosystem. Brad explores AWS's new native vector support for DynamoDB and explains why placing vectors inside operational databases slashes latency and infrastructure costs. Mitch shares top takeaways from Black Hat, focusing on "Shift Zero"—the practice of governing developer intent at the point of origin rather than catching errors downstream. Finally, the duo discusses prompt fatigue, non-deterministic pipelines, and how tools like Tencent's DB Agent Memory might help systems clean up their own technical debt. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#33
August 13 · 44 min
Stop Reviewing Agent Code? Anthropic, DynamoDB & AI Tech Debt | Agents of Dev Episode 33
Email your questions and topics to the podcast here. Are your AI agents creating more technical debt than they actually solve? In this episode of Agents of Dev, hosts Mitch Ashley and Brad Shimmin break down the fast-moving world of modern AI development, starting with AWS DynamoDB adding native vector support to optimize query latency. They dig into Black Hat insights like "Shift Zero" intent governance, why Anthropic is telling developers to strip down their prompt instructions, and how human alarm fatigue makes reviewing agent tool calls almost pointless. Finally, they wrap up with "The Drop," highlighting Tencent's graph-based agent memory hub and exploring how sci-fi clones from Dune mirror the current state of AI memory. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#32
August 6 · 46 min
Unifying Data Stacks, Open-Weight Controversies & Google's Anti-Gravity CLI | Agents of Dev Episode 32
Email your questions and topics to the podcast here. As enterprise AI stacks grow increasingly complex, the race is on to simplify data pipelines, unify developer work surfaces, and tame ballooning token costs. In this episode of Agents of Dev, hosts Mitch Ashley and Brad Shimmin break down the biggest shifts across database architecture, agentic tools, and AI infrastructure. They explore RegattaDB's unified engine for OLTP, analytical, and vector workloads, Atlassian's expanding AI lifecycle strategy, and Mitch's eight-layer AI Stack framework highlighting why "presence is not completeness." They also discuss the industry drama surrounding Moonshot AI's 2.8-trillion parameter Kimi K3 open-weight model, Anthropic’s distillation accusations, and the economic shift toward local models. To close out, Brad shares key takeaways from migrating to Google's Anti-Gravity CLI and what new agent workspace protocols mean for future developer workflows. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#31
July 30 · 37 min
Securing AI Agents: Snowflake CSTO Mayank Upadhyay on Cortex AI Gateway Presented by Snowflake | Agents of Dev Episode 31
Email your questions and topics to the podcast here. As enterprise AI evolves from simple chatbots into autonomous agents, traditional security methods are no longer enough to protect your company's crown jewels. In this episode of Agents of Dev presented by Snowflake, hosts Mitch Ashley and Brad Shimmin sit down with Mayank Upadhyay, Chief Security and Trust Officer at Snowflake, to dive into the security and governance shift required for agentic workflows. Mayank explains why static API keys fail in non-deterministic environments, how workload identity gives autonomous agents true accountability, and how Snowflake’s Cortex AI Gateway centralizes model routing, cost management, and Model Context Protocol (MCP) tool governance. This and more on Agents of Dev, part of The Futurum Group Podcast Network. This episode was presented by Snowflake.
#30
July 23 · 48 min
How Red Team/Blue Team AI, JSpace, & Agentic IDEs Are Changing Development | Agents of Dev Episode 30
Email your questions and topics to the podcast here. What happens when you pit Google Gemini against Anthropic Claude in a "red team/blue team" duel to fact-check AI research, or peek inside an LLM’s subconscious "Jacobian Space"? In this episode of Agents of Dev, analysts Mitch Ashley and Brad Shimmin explore the rapidly evolving landscape of AI engineering and how modern tools are turning everyone into a builder. They cover IBM’s multi-agent IDE update ("Bob"), AWS OpenSearch’s jump into real-time observability, and Anthropic’s groundbreaking research into model introspection. Plus, they dive into the "Drop" segment to discuss practical developer breakthroughs like Accordion for smart agent memory compaction, Cactus for ultra-fast micro-model function calling, Herder for terminal-based agent management, and Prefect's acquisition of Dagster Labs. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#29
July 16 · 44 min
Inside Google’s OKF, Linux Appia, and Ornith 1.0 Self-Scaffolding Models | Agents of Dev Episode 29
Email your questions and topics to the podcast here. Can autonomous AI agents finally be trusted to modify your enterprise code bases and databases without a human handholding them? In this episode of Agents of Dev, analysts Mitch Ashley and Brad Shimmin break down the explosive evolution of the agentic control plane. From the newly launched Linux Appia Foundation mapping out AI conformity to Google's Open Knowledge Format (OKF) operationalizing self-organizing LLM wikis, the core infrastructure for developer workflows is shifting fast. The duo also dives deep into the general availability of AWS's DevOps agent, Oracle's structural thesis on database-level security, and hands-on impressions of Ornith 1.0—the open-source, self-scaffolding model shaking up agentic coding. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#28
July 9 · 34 min
Databricks LTAP Collapsing the Wall Between OLAP and OLTP | Agents of Dev Episode 28
Email your questions and topics to the podcast here. AI agents are completely transforming how enterprise data moves, but running them shouldn't break your token budget. In this episode of Agents of Dev, practice leads Mitch Ashley and Brad Shimmin break down the massive infrastructure shifts coming out of Microsoft Build and Snowflake Summit. Discover how tools like Microsoft Fabric IQ, Project Rayfin, and Snowflake's "Coco" (Cortex Code) ecosystem are elevating metadata into actionable ontologies, shrinking token counts, and cutting operational costs. They also discuss Anthropic’s unsettling Claude 4.8 self-awareness disclosures, evaluates Google’s incredibly fast, encoder-free Gemma 4 12B local model, and extracts a crucial software design lesson from Blue Origin’s recent launchpad failure. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#27
July 2 · 49 min
Agentic Reality Check: Multi-Cloud Data, Zero Trust Security, and the Death of ETL | Agents of Dev Episode 27
Email your questions and topics to the podcast here. AI is moving faster than ever, but enterprise leaders are discovering a tough reality: deploying autonomous agents is not just about choosing the right model. It is about building the right control systems. In this episode of Agents of Dev, practice leads Mitch Ashley and Brad Shimmin reflect on a busy conference season covering Snowflake, Databricks, and AWS. They look beyond the keynote announcements to explore why companies are moving away from complex ETL pipelines and letting data stay where it already lives. They also discuss the rapid rise of the Model Context Protocol (MCP), how open source AI frameworks are becoming commercial products, and why Zero Trust security must evolve for AI agents that make unpredictable decisions. From managing AI costs with FinOps to deciding where enterprise data belongs, this episode offers practical insights for building secure, reliable, and people-focused AI environments. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#26
June 18 · 41 min
Why Claude Opus 4.8 Changes Everything: Models as Platforms, AI Microgravity & Graphify | Agents of Dev Episode 26
Email your questions and topics to the podcast here. Is your AI development stack quietly trapping you in a state of "microgravity" vendor lock-in? In this episode of Agents of Dev, hosts Mitch Ashley and Brad Shimmin break down why the latest wave of AI engines—headlined by Anthropic’s Claude Opus 4.8—are evolving into fully integrated operational systems rather than standalone models. Brad reveals how he used Claude Code to blueprint a complex 2,800-line Python migration in just five turns, shifting the architecture from a simple lift-and-shift to an optimized, AI-forward upgrade. The duo also dives into Starburst’s decentralized "catalog of catalogs" approach to data gravity, Harness's savvy acquisition of CodeCov to eradicate verification debt at the point of origin, and a game-changing open-source tool called Graphify that spins up queryable context graphs for agentic workflows with a single terminal command. This and more on Agents of Dev, part of The Futurum Group Podcast Network. #ClaudeOpus #Anthropic #AgenticAI #Graphify #DevSecOps #DataEngineering #AIPlatform #SoftwareArchitecture #AgentsOfDev
#25
June 11 · 1 hr
AI Agents Are Breaking the Stack: Fabric IQ, Snowflake Coco, & Gemma 4 | Agents of Dev Episode 25
Email your questions and topics to the podcast here. AI agents are completely transforming how enterprise data moves, but running them shouldn't break your token budget. In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin break down the massive infrastructure shifts coming out of Microsoft Build and Snowflake Summit. Discover how tools like Microsoft Fabric IQ, Project Rayfin, and Snowflake's "Coco" (Cortex Code) ecosystem are elevating metadata into actionable ontologies, shrinking token counts, and cutting operational costs. They also discuss Anthropic’s unsettling Claude 4.8 self-awareness disclosures, evaluates Google’s incredibly fast, encoder-free Gemma 4 12B local model, and extracts a crucial software design lesson from Blue Origin’s recent launchpad failure. This and more on Agents of Dev, part of The Futurum Group Podcast Network. #AgentsofDev #AIAgents #GraphDatabases #DORAReport #AIFinOps #TokenEconomics #RedHatSummit #SAPSapphire #DevOps
#24
May 28 · 50 min
The Truth About 10x AI Productivity plus Google Anti-Gravity 2.0 & MCP Washing | Agents of Dev Episode 24
Email your questions and topics to the podcast here. Google is rebuilding its ecosystem around Anti-Gravity 2.0 and enterprises shifting toward smaller, localized models like Gemma 4. In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin break down the reality behind the "10x AI productivity" hype and explore why the industry is shifting toward "token economics" cost management. The duo analyzes the massive announcements from Google I/O, including Gemini 3.5 Flash, Gemma 4 open-weight models, and the evolution of Anti-Gravity 2.0 into a vertically integrated, multi-surface agentic platform. They also tackle the trend of "MCP washing" following Model Context Protocol server launches from Alteryx and Informatica, look at Dell's new desk-side agentic hardware, and share hands-on strategies for parallel multi-stream LLMs and localized graph architectures using DuckDB and pandas. This and more on Agents of Dev, part of The Futurum Group Podcast Network.
#23
May 21 · 1 hr
The Real Cost of AI Agents: Graph Databases, DORA J-Curve, and AI FinOps | Agents of Dev Episode 23
Email your questions and topics to the podcast here. As enterprises rush to deploy autonomous AI agents, many are waking up to tens of thousands of dollars in unexpected API bills, proving that the shift to agentic AI is as much an economic challenge as it is a technical one. In this episode of the Agents of Dev podcast, Mitch Ashley and Brad Shimmin break down the massive architectural shifts hitting the software development lifecycle, highlighting why this has officially become the year of the graph database for agentic context. They dive deep into the latest DORA report to explore the infamous "J-curve" of AI productivity, explaining why "reclaimed capacity" is the true ROI metric organizations must track. Finally, Mitch and Brad share boots-on-the-ground recaps from SAP Sapphire and Red Hat Summit, unpacking everything from the "metal-to-agent" infrastructure stack and open-weight models to breakthrough AI FinOps tools like Revenium and G-brain built to tame runaway token economics. This and more on Agents of Dev, part of The Futurum Group Podcast Network. #AgentsOfDev #AIAgents #GraphDatabases #DORAReport #AIFinOps #TokenEconomics #RedHatSummit #SAPSapphire #DevOps
#22
May 14 · 58 min
Is RAG Dead? Graph RAG, Rogue Agents, and the Evolution of the AI Development Stack | Agents of Dev Episode 22
Email your questions and topics to the podcast here. What happens when your AI agent decides to delete your production database because it was "pretty sure" it wasn't necessary? In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin dive deep into the shifting landscape of enterprise AI, starting with why "naive RAG" is failing practitioners and how Graph RAG and context engines (like the newly unveiled Loveless AI) are providing the deterministic context businesses actually need. They analyze a viral cautionary tale of a rogue AI agent that deleted a production database, highlighting the desperate need for an Agent Control Plane and robust governance. The duo also explores the evolution of the developer's "work surface," from the agentic shifts in Cursor 3.2 to the surprisingly powerful enterprise productivity of Amazon Q. Whether you're building data estates with DuckDB or navigating the vertically integrated stacks of Google and AWS, this episode covers the "art of the doable" in the age of agentic development. This and more on Agents of Dev, part of The Futurum Group Podcast Network. #AgentsofDev #AI #GenerativeAI #AgenticDev #SoftwareEngineering #RAG #GraphRAG #AgentsOfDev #DevOps #AIInfrastructure #DataEngineering #AmazonQ #CursorIDE
#21
May 7 · 45 min
Is the Data Lakehouse Dead? Databricks, Iceberg, & Google's New AI Stack | Agents of Dev Episode 21
Email your questions and topics to the podcast here. The "open table format wars" are officially over, but a new era of vertically integrated AI stacks is just beginning. In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin dive deep into the massive wave of announcements from Google Cloud Next, focusing on the shift from systems of intelligence to "systems of action." Brad breaks down why Databricks embracing Apache Iceberg v3 is a game-changer for data portability and how Google’s new Agentic Data Cloud—powered by Spanner Graph and a new "Knowledge Catalog"—is turning raw object storage into self-enriching reasoning engines. Meanwhile, Mitch puzzles over Elon Musk’s SpaceX/Cursor acquisition and proposes a new framework for the emerging AI stack. From "vibe coding" your way to Mars to the controversial rise of the Model Context Protocol (MCP) as the universal access layer, this conversation explores how the industry is optimizing through abstraction to give AI true agency. This and more on Agents of Dev, part of The Futurum Group. #GoogleCloudNext #AI #DataEngineering #GenerativeAI #Databricks #CloudComputing #TechPodcast #AgenticAI #OpenSource #DataLakehouse
#20
April 30 · 44 min
Mainframe MCP, YOLO AI Workflows, and the Mythos Security Crisis | Agent of Dev Episode 20
Email your questions and topics to the podcast here. In a world where AI-driven security can find a million vulnerabilities in minutes, "business as usual" for developers is officially dead. In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin of The Futurum Group dive into the rapid evolution of the software lifecycle. They explore how Model Context Protocol (MCP) is revitalizing the mainframe by bridging the skills gap and acting as the "can opener" for legacy systems. Brad shares his journey of going "Full YOLO" with agentic development in the Signal project, explaining why high-order reasoning is changing how we maintain complex Python codebases. The duo also tackles the "Grep vs. RAG" debate, examining why simplicity and determinism often beat complex semantic searches. Finally, they discuss the security fallout of Anthropic’s Mythos model and look to the stars to discuss the "silent failure" engineering of NASA’s Artemis II mission. This and more on Agents of Dev, part of The Futurum Group. #AgentsofDev #AgenticAI #DevOps #Mainframe #CyberSecurity #SoftwareEngineering #MCP #RAG #ArtemisII #GenerativeAI #TechPodcast
#19
April 23 · 35 min
Oracle Mission Critical Scale, AI Benchmarking Pitfalls & Cost of Context | Agents of Dev Episode 19
Email your topics and questions to the podcast here. What if the biggest challenge in AI is not building it, but understanding how it changes underneath you? In this episode of Agents of Dev, Mitch Ashley and Brad Shimmin explore Oracle’s latest mission critical data infrastructure advancements alongside the fast evolving realities of AI systems, benchmarking, and cost management. The discussion covers Oracle’s push toward extreme scale and high availability for global workloads, while also unpacking the hidden instability in AI systems caused by shifting APIs, model updates, and opaque pricing changes. They also discuss why transparency is becoming essential for trust, how benchmarking AI models can mislead if not carefully designed, and why developers are increasingly relying on memory management, caching, and local models to control cost and performance. The episode also highlights creative engineering approaches, from simple tools like curl to local AI experimentation, showing how practitioners are adapting to a rapidly changing AI landscape. This and more on Agents of Dev. #AgentsofDev #AI #ArtificialIntelligence #DataManagement #CyberSecurity #Microsoft #Anthropic #GraphDatabase #MachineLearning #AIGovernance

Showing 1–20 of 26 episodes