AI agents need cheaper sandboxes | Alex Jung from Unikraft (Heavybit DevGuild)

In this episode, Siddhartha Borah, cofounder and CTO of Brackett AI, joins us at Heavybit DevGuild. Siddhartha explains how Brackett is building agents that can learn complex business work — especially in areas like supply chain and operations, where the real context often lives in people’s heads rather than documentation. We talk about teaching agents by doing the work, turning messy human knowledge into reliable workflows, agent-in-the-loop systems, memory, knowledge graphs, and why the last mile of enterprise automation is so hard.Links:• Siddhartha Borah’s LinkedIn• Brackett AI• Heavybit DevGuild 00:00 Intro00:30 What Brackett AI is building01:40 Learning work that was never written down03:30 Turning human context into repeatable workflows05:20 Why real-world agents fail without hidden context08:00 Teaching agents by doing the work09:50 Handling edge cases and human handoff11:30 Memory, knowledge graphs, and “déjà vu” for agents14:00 Where to learn more

In this episode, James Arthur, cofounder and CEO of ElectricSQL, joins us at Heavybit DevGuild. James explains why Electric has evolved from Postgres sync into data infrastructure for collaborative multi-agent systems. We talk about Durable Streams, Electric Agents, why agents should be modeled as long-lived data rather than compute, and why the future of agent infrastructure may look more like sync than sandboxes.Links:• James Arthur’s LinkedIn• James Arthur on X• Electric• Electric blog• Heavybit DevGuild 

In this episode, Dave Fletcher, cofounder of LeadDev, joins us at LDX3 London.Dave shares what LeadDev is seeing from thousands of engineering leaders, including what tools they are buying, how AI coding tools are changing downstream needs, and why observability, CI/CD, testing, reliability, and security are rising up the priority list.We also talk about how DevTools companies should message AI without triggering engineers’ bullshit detectors, why specificity beats hype, and why in-person events are becoming more important as digital channels get crowded.Links:Dave Fletcher’s LinkedInLeadDevLDX3 LondonLeadDev events

In this episode, Dennis Pilarinos, founder and CEO of Unblocked, joins us at LeadDev in London.Dennis explains why AI coding agents are only as good as the context they can access, and why every new agent is like a developer on their first day at a company.We talk about organizational context, token savings, MCP, stale documentation, knowledge graphs, permissions, identity across systems, and why the next big infrastructure layer for AI development may be the context layer.LinksDennis Pilarinos on LinkedInDennis Pilarinos on XUnblockedUnblocked BlogContext EngineeringLeadDevLDX3 London

In this episode, Ian Coe, Co-founder and CEO of Tonic.ai, joins us at Heavybit DevGuild.Ian explains how Tonic started by helping developers create safe, realistic data for testing and development, and how that has expanded into AI model training, de-identification, and synthetic datasets for reinforcement learning.We discuss structured vs. unstructured data, healthcare and financial services use cases, Tonic Fabricate, Tonic Textual, fine-tuning, RL environments, and why enterprises are becoming more interested in training models on their own data.LinksIan Coe's LinkedInTonic.aiTonic FabricateTonic TextualTonic BlogHeavybit DevGuild

In this episode, Alex Jung, cofounder and CPO of Unikraft, joins us at Heavybit DevGuild. Alex explains how Unikraft went from an academic unikernel project to a cloud platform for running highly efficient workloads. We talk about why unikernels can make applications faster, cheaper, and more secure, how AI agents are driving demand for sandboxed compute, and why the next wave of software generation may require a much more efficient cloud infrastructure layer.Links:• Alex Jung’s LinkedIn• Alex Jung on X • Unikraft• Unikraft on GitHub • Unikraft open source project 

In this episode, Kim Maida joins the podcast to talk about DevRel, AI, developer experience, and measuring what actually matters. We also dive into her developer empowerment flywheel and how Keycard is approaching identity and access for AI agents.Links:   •  Kim Maida   •  Keycard   •  Kim Maida on LinkedIn   •  Kim Maida on X

In this episode, Robby Russell joins to talk about how Oh My Zsh went from a tool for a few coworkers to one of the most popular open source developer tools in the world. They cover developer experience, open source maintenance, AI-assisted pull requests, and how AI could reshape software consulting.Links:   •  Robby Russell   •  Oh My Zsh   •  Maintainable podcast   •  Ruby on Rails Podcast   •  Planet Argon

In this episode, Swyx, founder of AI Engineer, joins us live from AI Engineer Europe in London, alongside Louis Knight-Webb. We cover how AI Engineer grew from a single San Francisco conference into a global community, why industry AI work needs better venues for sharing, and what is changing in AI DevTools, code execution, research, PLG, enterprise sales, and human-written content in the age of AI.Links:• Swyx's LinkedIn• Swyx on X• Louis Knight-Webb's LinkedIn• Louis Knight-Webb on X• AI Engineer• AI Engineer Europe• AI Engineer YouTube• AI Engineer on X

In this episode, Dan Moore from FusionAuth breaks down how the company integrated Permify after the acquisition. We talk about customer communication, pricing and packaging, migration planning, internal enablement, and the practical work that turns an acquisition into a successful product integration.Links:   •  FusionAuth   •  Permify   •  Dan Moore on Bluesky