Edith Harbaugh: Code Is Cheap. The Last Mile Isn't.

Automated software delivery is moving faster than the people who used to approve every release. LaunchDarkly co-founder and CEO Edith Harbaugh tells Guy Podjarny why rollback should be autonomous, why one-shot coding agents fail, and why the bottleneck has moved to the last mile.What we cover:– How automated software delivery and autonomous rollback take humans out of the 2 a.m. incident– What makes a good feature flag, and when a homegrown system stops being enough– Why coding agents need an explicit definition of success before they write a line– Where release management fits in an AI software factory– Feature flags for AI agents: switching models and prompts at runtime– Running a company in the agentic coding era, and why teams should take dumb betsChapters:00:00:00 - Introduction00:02:16 - From quarterly releases to 30,000 changes a day00:03:32 - Autonomous rollback and automated software delivery00:06:25 - What makes a good feature flag00:10:46 - Defining success for coding agents00:16:42 - Agent fleets and engineers as reviewers00:24:05 - Feature flags in the software factory00:29:02 - The bottleneck moves to the last mile00:39:24 - Runtime controls for AI agents00:44:26 - Running a company in the agentic coding eraBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/e0p🔔 Subscribe for weekly episodes on AI-native developmentWhere has the bottleneck moved in your team's automated software delivery? Tell us in the comments.

Agentic coding has outgrown the text box. Maggie Appleton, staff research engineer at GitHub Next, argues that walls of CLI output and AGENTS.md files serve coding agents far better than the humans steering them. She walks through Chopin, the GitHub Next prototype where background agents turn a team's planning chat into a structured plan, and makes the case for agents that prompt you instead of waiting to be prompted.What we cover:– Why text-heavy agentic coding tools favor agents over humans– Boundary objects: what AGENTS.md, skill files and MCP servers get right and wrong– Multiplayer planning with coding agents in GitHub Next's Chopin– Human-to-agent handoffs, formal verification and spec-driven development– Why software quality is rising, not falling, with coding agents– Which roles still matter: PMs, designers and engineersChapters:00:00:00 - Introduction00:02:14 - Inside GitHub Next and the Chopin project00:05:57 - Why solo agents break team alignment00:07:52 - Why coding agents struggle with design00:12:16 - Boundary objects: AGENTS.md, skills and MCP00:18:27 - Planning with coding agents as collaborators00:26:49 - Multiplayer planning with background agents00:31:57 - Agent handoffs and spec-driven development00:37:41 - Quality, craft and who decides what to build00:44:29 - Starting a career in agentic codingBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/j3d🔔 Subscribe for weekly episodes on AI-native developmentShould coding agents be prompting you more? Tell us in the comments how your team plans agentic coding work today.

AI attackers now move at machine speed, while most security teams still move at meeting speed. Former Robinhood CSO Caleb Sima has seen an AI attacker up close, and he argues the only way to keep up is to take humans out of the defense loop.Caleb joins Guy Podjarny to unpack what changes when the attacker is an agent: one that hits every surface at once, reverses open-source code to find a 0-day, and once inside, burns your own AI tokens to keep going.What we cover:– Why AI attackers outran the red teamers' predictions, including 400 Linux 0-days found at 2.5 hours each– Whether sandboxes can contain AI agents, and why external monitoring matters– Who is accountable when your AI agent hacks someone else– What actually changes when an AI attacker gets inside your company– Deception, digital twins and forensic-grade detection– Automating security: instant patching, one-way vs two-way doors, and security teams that ship the fixChapters:00:00:00 - Introduction00:02:58 - AI security goes mainstream00:05:52 - Why AI attackers surprised the red teamers00:12:50 - Can sandboxes contain AI agents?00:18:03 - Security vs safety vs alignment00:22:18 - Who's accountable when your AI agent hacks?00:29:00 - What changes when an AI attacker hits you00:36:47 - Detection, forensics and deception00:44:10 - Machine speed vs meeting speed: automating defense00:48:53 - Security teams that ship the fixBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/9ja🔔 Subscribe for weekly episodes on AI-native developmentAre your defenses ready for AI attackers, or still running at meeting speed? Tell us in the comments.

Your coding agent should never hold a credential, and in a software factory that rule is only the start. Recorded live at the AI Security Summit in San Francisco, Simon Maple sits down with Eli Aleyner (Docker), Marcelo Sousa (Snyk) and Ian Livingstone (Keycard), co-authors of Agent Baseline, a security reference architecture for the AI software factory.The panel unpacks why agents have to be treated as fundamentally unpredictable actors, why sandboxing alone is not enough when skills and MCP servers make the supply chain dynamic, and why approval prompts collapse into fatigue. Marcelo's verdict: sometimes these are kids with chainsaws.What we cover:– The three levels of autonomy, from coding agents to fully agent-directed workflows– Sandboxing coding agents and injecting secrets at the egress point– Why the agent supply chain is dynamic: skills, MCP servers and cloned repos– Prompt injection, least privilege and agents monitoring other agents in the harness– Approval fatigue: YOLO mode vs CYA mode– Where to start securing your software factory on Monday morningChapters:00:00:00 - Introduction00:03:14 - What Agent Baseline is and who it's for00:04:17 - The three levels of software factory autonomy00:06:53 - Sandbox boundaries and manufacturing trust in agents00:09:22 - Why coding agents should never hold credentials00:11:44 - The dynamic agent supply chain: skills and MCP servers00:14:57 - Untrusted input, least privilege and agents watching agents00:18:02 - Approval fatigue: YOLO mode vs CYA mode00:20:22 - Are poisoned agent skills a supply chain problem?00:22:45 - Where the panel disagreed, and where to start MondayBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/gsh🔔 Subscribe for weekly episodes on AI-native developmentWhich level of the software factory are you at today, and what are you locking down first? Tell us in the comments.

Skills and MCP servers are the new npm, and AI coding agents are pulling them into enterprise codebases faster than security can keep up. Manoj Nair, Chief Innovation Officer at Snyk, joins Guy Podjarny at the Snyk AI Security Summit to map where enterprise agentic coding really stands.They get into why a 50,000-agent Claude Code rollout was pulled forward by weeks, why the autonomous attacker has some CISOs ready to accept an outage to patch faster, and how security is moving into the agent loop of the software factory.What we cover:– Why leadership and culture, not tooling, decide how fast enterprises adopt AI coding agents– The three risks of agentic coding: insecure output, the skills and MCP server supply chain, and agent behavior– Running coding agents locked down: sandboxes, agent identities and no open internet– Who owns the code an agent writes, and why AI-native organizations treat it as shared risk– Securing the software factory: security context in the agent loop and agentic remediation of the backlog– What still has to be solved before enterprises trust agents: identity, intent and behaviorChapters:00:00:00 - Introduction00:01:50 - The state of enterprise AI and agent adoption00:09:45 - Leadership, culture and a 50,000-agent rollout00:11:18 - Agentic coding risks: output, supply chain, behavior00:14:29 - Locked-down coding agents and the security trade-off00:18:01 - Who owns agent output? Compliance and accountability00:22:54 - The autonomous attacker and security technical debt00:27:56 - Securing the product vs securing the software factory00:33:00 - Agentic remediation and the security backlog00:40:18 - What enterprises need to trust AI coding agentsBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/jw0🔔 Subscribe for weekly episodes on AI-native developmentWhere is your organization on the path from copilot to AI coding agents? Tell us in the comments.

Agent feedback loops are how a software factory maintains itself: agents mining their own session logs, PR comments and production signals, then fixing what they find. Maria Gorinova, member of technical staff at Tessl, explains how to build them, and why vague intent turns a loop into a mess.What we cover:– What a software factory is: repeatable workflows, a shared execution environment and agent feedback loops– Why agent context alone isn't enough, and where harness engineering comes in– The three stages of a feedback loop: analyze a data stream, find the problem, request the fix– Inner, middle and outer loops, from verifiers on every PR to self-healing in production– How Tessl's "comment diet" loop catches drift that no single PR causes– Operators and stewards: the human roles in a software factoryChapters:00:00:00 - Introduction00:02:36 - What is a software factory?00:05:38 - From agents to context to harness engineering00:09:15 - The three stages of an agent feedback loop00:13:16 - Inner, middle and outer feedback loops00:18:35 - Why vague intent breaks agent feedback loops00:21:50 - Comment bloat, drift and the rule of three00:25:48 - The software factory mindset00:27:45 - Loops, feedback loops and meta feedback loops00:32:45 - Operators, stewards and building your first loopBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/2uf🔔 Subscribe for weekly episodes on AI-native developmentWhich agent feedback loop would you build first in your own software factory? Tell us in the comments.

AI coding agents now write almost all of Legora's code, with engineers running up to ten in parallel. Yet the CTO behind that shift thinks measuring adoption by tokens burned is a mistake. Jacob Lauritzen, CTO of Legora, joins Simon Maple to unpack what actually scales when agents do the typing, and where human judgment still has to stay in charge.What we cover:– How Legora's engineers orchestrate many AI coding agents at once, from shared local dev resources to cloud agent setups– An agentic loop that reproduces, fixes and tests bugs straight from Slack– Why token maxing is a bad way to drive AI adoption, and what to measure instead– Agent evaluation: regression testing the harness as models and prompts change every week– Decision models, the verifier's rule, and where autonomy ends and human judgment begins– Why AI agents need more than a chat interfaceChapters:00:00:00 - Introduction00:01:47 - Legora's growth from $100M to $200M ARR00:03:44 - Parallel coding agents and bug-fix loops00:07:15 - Senior engineers vs AI-first engineers00:10:10 - Why token maxing is the new lines of code00:14:13 - What AI can and can't verify in legal work00:17:30 - Decision models vs autoregressive LLMs00:24:22 - Weekly model benchmarks and harness evals00:26:58 - Why AI agents need more than a chat interface00:35:03 - The verifier's rule and litigation strategyBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/lpg🔔 Subscribe for weekly episodes on AI-native developmentIs token count telling you anything real about how your team uses AI coding agents? Tell us in the comments.

Honeycomb went from 30 to 70 merged pull requests a day in three months. The catch: automated code review, not code generation, became the real bottleneck of software automation. Liz Fong-Jones, Technical Fellow at Honeycomb, explains why incidents still rose 1.5x, how their internal bot Autobot now reviews every PR, and why AI amplifies whatever org you already have.What we cover:– How automated code review lets humans focus on design, not trivial bugs– Using a decision model like Jev to decide which PRs are safe to auto-merge– What makes a codebase ready for AI coding agents– When to trust AI agents with production incidents, and when they're just throwing darts– Why "Claude did it" isn't an excuse, and what ownership means with AI– How open source maintainers can handle a flood of AI slop pull requestsChapters:00:00:00 - Introduction00:06:41 - Why AI amplifies dysfunctional engineering orgs00:10:45 - What makes a codebase AI ready00:14:01 - Honeycomb's Autobot and automated code review00:19:06 - Using Jev to decide which PRs are safe to merge00:26:51 - Trusting AI agents during production incidents00:30:40 - Least privilege and guardrails for coding agents00:33:25 - If your name's on it, you own it00:38:55 - AI slop pull requests and open source00:45:56 - Will observability engineering survive as a role?Build your software factory, one workflow at a time, with Tessl:https://tessl.co/nlr🔔 Subscribe for weekly episodes on AI-native developmentIs your team's review capacity keeping up with your AI coding agents? Tell us in the comments.

HumanLayer CEO Dexter Horthy on why his team's bet on spec driven development nearly wrecked their own codebase, and what changed his mind about reviewing AI-written code. Dexter walks through the "YOLO pull request" experiment that ran for months before the codebase became unusable, the "dumb zone" that limits how much context you can hand a model, and why he now believes there will always be alpha in reviewing something.What we cover:– How HumanLayer's "read the plan, skip the code" experiment quietly broke their own codebase– Why context windows have a "dumb zone," and what that means for context engineering– Sean Grove's idea that specs, not code, are becoming the durable artifact– Building overnight agents that review and fix code before a human ever sees it– Why treating software development like a factory changes how bugs compound– What Dexter thinks his advice on reviewing AI-written code will look like in 12 monthsChapters:00:00:00 - Introduction00:02:15 - Meet Dexter Horthy, CEO of HumanLayer00:06:03 - The "dumb zone": why more context makes models dumber00:06:40 - Sean Grove's "the spec is the new code"00:09:22 - The YOLO pull-request experiment that broke their codebase00:11:09 - Letting the model own the architecture00:17:03 - Planning as expected-pain management00:28:15 - Slop Code Bench and the maintainability oracle problem00:37:59 - Why software factories aren't like car factories00:42:49 - "There will always be alpha in reviewing something"Build your software factory, one workflow at a time, with Tessl:https://tessl.co/4ep🔔 Subscribe for weekly episodes on AI-native developmentWhere do you land on the determinism-to-adaptability slider — plan tightly, or let the agent run? Tell us in the comments.

Spec-driven development is reshaping what software engineers actually do all day, and Marc Brooker, VP and Distinguished Engineer at AWS, has read 3,000 to 4,000 postmortems that convinced him the code was never the hard part. In this episode of The AI Native Dev, Marc explains why testing and specification are now the real engineering work, why metastable failures keep taking down systems that look perfectly healthy, and why he still won't let AI write a single word of his blog.What we cover:– Why spec-driven development is turning testing into the hardest part of software– What metastable failures are, and why they keep taking healthy-looking systems down– How agentic coding tools can learn from postmortems and build their own memory– Why classic authorization breaks down once you're writing agentic policy– Why Marc won't let AI write his blog, but is fully comfortable with AI-generated code– What it takes to bring junior engineers up to speed in an AI-native industryChapters:00:00:00 - Introduction00:02:17 - Marc Brooker: 18 years at AWS building agentic dev tools00:04:36 - Inside Strands, AWS's open source agent SDK00:06:44 - Fifteen years on call, and what agents still can't debug00:11:23 - Metastable failures and the humility of 4,000 postmortems00:14:32 - Teaching agents to learn from postmortems and build their own memory00:20:14 - Why agentic policy needs more than classic authorization00:32:14 - The agentic software development hypothesis: spec-driven development, oracles, and testing00:39:56 - Why Marc won't let AI write his blog (but will let it write code)00:46:37 - Advice for leveling up junior engineers in the AI eraBuild your software factory, one workflow at a time, with Tessl:https://tessl.co/[SLUG]🔔 Subscribe for weekly episodes on AI-native developmentDo you trust an agent to run its own on-call rotation yet? Tell us in the comments.