GitLab and the Evolution of DevSecOps
Episode
56 min
Read time
2 min
Topics
Relationships, Artificial Intelligence, Software Development
AI-Generated Summary
Key Takeaways
- ✓Orbit Knowledge Graph vs. RAG: Rather than using retrieval-augmented generation, GitLab's Orbit indexes all SDLC data into a graph stored in ClickHouse, letting agents write Cypher queries on the fly to traverse relationships across 50,000+ repositories. This enables instant dependency audits — finding every repo using a vulnerable package version in seconds, not hours.
- ✓Composite Agent Identity for Auditing: GitLab tracks every agent action using a composite identity model: a service account representing the agent paired with the initiating human user. Every tool call, model response, and approval is logged in a session trace, letting enterprises reconstruct exactly which user triggered which agent action months after the fact.
- ✓No-Touch Bug Resolution Pipeline: GitLab's internal engineering teams run an automated triage agent that, upon receiving a bug report, reproduces the issue in a VM, screen-records the reproduction, attempts a fix, screen-records the result, and posts everything to a merge request — so engineers arrive to a ready-to-review fix rather than a raw bug report.
- ✓Air-Gapped AI Deployment Architecture: Enterprises running fully disconnected environments can self-host both a GitLab instance and a dedicated AI gateway, routing requests to on-premise models without any external network calls. This eliminates the latency of cloud round-trips and satisfies data sovereignty requirements in government, automotive, and high-security financial sectors.
- ✓Breaking Change Policy for Long-Lived Deployments: Because some enterprise customers run GitLab versions that are years old, GitLab enforces a policy of one major version per year, with breaking changes announced a minimum of twelve months in advance. Engineers must treat each release as a package that may not reach a customer's machine for four or more months.
What It Covers
GitLab VP of Engineering Tim Zalman discusses how GitLab's DevSecOps platform approaches the AI agent era, covering the Duo Agent Platform, the Orbit knowledge graph, air-gapped deployments, composite agent identity for enterprise governance, and how the competitive landscape is shifting as coding tools like Cursor enter code hosting.
Key Questions Answered
- •Orbit Knowledge Graph vs. RAG: Rather than using retrieval-augmented generation, GitLab's Orbit indexes all SDLC data into a graph stored in ClickHouse, letting agents write Cypher queries on the fly to traverse relationships across 50,000+ repositories. This enables instant dependency audits — finding every repo using a vulnerable package version in seconds, not hours.
- •Composite Agent Identity for Auditing: GitLab tracks every agent action using a composite identity model: a service account representing the agent paired with the initiating human user. Every tool call, model response, and approval is logged in a session trace, letting enterprises reconstruct exactly which user triggered which agent action months after the fact.
- •No-Touch Bug Resolution Pipeline: GitLab's internal engineering teams run an automated triage agent that, upon receiving a bug report, reproduces the issue in a VM, screen-records the reproduction, attempts a fix, screen-records the result, and posts everything to a merge request — so engineers arrive to a ready-to-review fix rather than a raw bug report.
- •Air-Gapped AI Deployment Architecture: Enterprises running fully disconnected environments can self-host both a GitLab instance and a dedicated AI gateway, routing requests to on-premise models without any external network calls. This eliminates the latency of cloud round-trips and satisfies data sovereignty requirements in government, automotive, and high-security financial sectors.
- •Breaking Change Policy for Long-Lived Deployments: Because some enterprise customers run GitLab versions that are years old, GitLab enforces a policy of one major version per year, with breaking changes announced a minimum of twelve months in advance. Engineers must treat each release as a package that may not reach a customer's machine for four or more months.
Notable Moment
Zalman describes customers who run GitLab on spinning hard drives rather than SSDs, then express surprise at poor performance under thousands of concurrent users — illustrating how self-managed software forces engineering teams to account for hardware configurations entirely outside their control.
Episode Transcript
AI coding agents can already reach well beyond the code in front of them. Through connections to tools like Slack, Jira, and Sentry, an agent can pull in issues, discussions, and production signals from across an organization. However, that context is scattered across many separate systems with no shared map of how it all connects and no common controls over what an agent is allowed to touch. For a large enterprise, stitching that together while staying compliant, auditable, and sometimes fully cut off from the internet is a hard problem. GitLab runs one of the major platforms where software gets built, hosted, and shipped. That gives it a wide view of everything around the code, which shapes how it is approaching the agent era. The bet is that AI is most useful when it can draw on the full context of a project and stay under enterprise control, rather than acting as a point tool bolted onto a single repository. Getting there means wrestling with the hard questions about governance, agent identity, and how much autonomy to hand over. Tim Zalman is the VP of Engineering at GitLab. In this episode, he joins Gregor Van to discuss bringing AI to the full software development lifecycle, why running agents under enterprise control matters for governance and auditing, how to track which human is behind an agent's actions, the realities of shipping air gapped software, and how the space is shifting as new players move into code hosting. Gregor Vand is a security focused technologist, having previously been a CTO across cybersecurity, cyber insurance, and general software engineering companies. He is based in Singapore and can be found via his profile at vand.hk or on LinkedIn. Hello, and welcome to Software Engineering Daily. My guest today is Tim Zalman. Thank you for joining, Tim. Thanks a lot for the invitation. Great to be here. Yeah. So you are the VP of engineering at GitLab, a company, a platform that I think a lot of our listener base will know about. For those that don't, we'll definitely cover that soon as well. But as we'd like to do, Tim, you've been at GitLab actually for quite a long time. I think nine and a half years at this point. But when I did look at your LinkedIn effectively, I did see you had a, I think a gaming company for a while before that. I'd love to hear just what that journey was doing that and then transitioning to GitLab. Yeah. Happy to do that. Yeah. It's quite a ride at GitLab. I'm here for nine and a half years. I've seen the company go from a 100 and... I think I was 112 or so employee, and now we are at 2,000 plus. And, yeah, my career goes back, I would say, a lot of different stations. I, in reality, started programming at the company of my parents. I learned especially coding from my mother. So this is where I started. I …
Get the full transcript (10,044 words) + summary by email — free
One-time email with the complete transcript and AI summary of this episode. No account needed.
One email, no spam. We’ll also show you what SignalCast does.
You just read a 3-minute summary of a 53-minute episode.
Get Software Engineering Daily summarized like this every Monday — plus up to 2 more podcasts, free.
Pick Your Podcasts — FreeKeep Reading
More from Software Engineering Daily
SED News: PHP’s Megabug, a Hidden Mic at OpenAI, and Dead AI Products
Nov 5 · 54 min
NVIDIA AI Podcast
NVIDIA’s Bartley Richardson on Why ‘Agentic AI Is Next-Level Automation’ - Ep. 258
May 28
More from Software Engineering Daily
Demis Hassabis and the Race to Build AI with Sebastian Mallaby
Nov 3 · 65 min
How I AI
Build your own company brain: the enterprise AI playbook from Stripe’s engineering team | Sharadh Krishnamurthy
Sep 7
Books, tools, and gear mentioned in this episode
SignalCast may earn commission on purchases via these links.
Tools
by GitLab
“GitLab VP of Engineering Tim Zalman discusses how GitLab's DevSecOps platform approaches the AI agent era, covering the Duo Agent Platform, the Orbit knowledge graph, air-gapped deployments”
by GitLab
“Orbit indexes all SDLC data into a graph stored in ClickHouse, letting agents write Cypher queries on the fly to traverse relationships across 50,000+ repositories”
by GitLab
“GitLab VP of Engineering Tim Zalman discusses how GitLab's DevSecOps platform approaches the AI agent era, covering the Duo Agent Platform, the Orbit knowledge graph, air-gapped deployments, composite agent identity for enterprise governance”
“Orbit indexes all SDLC data into a graph stored in ClickHouse, letting agents write Cypher queries on the fly”
“the competitive landscape is shifting as coding tools like Cursor enter code hosting”
More from Software Engineering Daily
We summarize every new episode. Want them in your inbox?
SED News: PHP’s Megabug, a Hidden Mic at OpenAI, and Dead AI Products
Demis Hassabis and the Race to Build AI with Sebastian Mallaby
AI Regulation and What Every Tech Company Should Know
Managing AI Coding Costs at Scale
Blitzy and Autonomous Software Development for Enterprise
Similar Episodes
Related episodes from other podcasts
NVIDIA AI Podcast
May 28
NVIDIA’s Bartley Richardson on Why ‘Agentic AI Is Next-Level Automation’ - Ep. 258
How I AI
Sep 7
Build your own company brain: the enterprise AI playbook from Stripe’s engineering team | Sharadh Krishnamurthy
The AI Breakdown
Sep 4
How AI Changed This Summer
How I AI
Sep 2
Grok Bot vs. OpenClaw: How I replaced my entire agent stack
Latent Space
Jul 28
Codex from 0 to 10M Users: Building ChatGPT Work — Akshay Nathan, OpenAI
Explore Related Topics
This podcast is featured in Best Cybersecurity Podcasts (2026) — ranked and reviewed with AI summaries.
Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.
You're clearly into Software Engineering Daily.
Every Monday, we deliver AI summaries of the latest episodes from Software Engineering Daily and 192+ other podcasts. Free for one show.
Start My Monday DigestNo credit card · Unsubscribe anytime