Arjun Mehta
Principal Engineer
Arjun Mehta is Principal Engineer at Glue, where he builds the AI systems that parse, analyze, and make sense of codebases at scale. Before Glue, Arjun spent 12 years as a backend and infrastructure engineer at companies ranging from early-stage startups to a publicly traded enterprise SaaS company. He's been the "bus factor = 1" engineer — the person who knew how the billing system worked, the person who got paged at 2am when the payment pipeline broke, the person new hires were told to "just ask Arjun" instead of reading documentation that didn't exist. He's also been on the other side: the new engineer who joined a team with 200K lines of undocumented code and no onboarding beyond "read the README and good luck." Arjun writes about the engineering side of the codebase intelligence problem — technical debt quantification, code health metrics, architecture patterns, dependency analysis, and why the tools engineers use today still leave massive knowledge gaps. His writing is technically rigorous but always connects back to the business impact, because he's learned (the hard way) that technical debt only gets addressed when leadership understands the cost.
The CTO's Guide to Product Visibility
How CTOs can answer board-level questions about features, competitive position, and technical health using code intelligence.
The Engineering Manager's Guide to Code Health
Everything engineering managers need to monitor, measure, and improve the health of their codebase.
Glue for Engineering Planning
Generate dev plans that show exactly which files to touch, what dependencies exist, and where the risks are.
Glue for Technical Debt Management
See where debt lives, who owns it, and what to fix first. Make tech debt visible to the entire product team.
Glue vs GitHub Copilot for Product Understanding
Copilot writes code. Glue reads it for the whole team. Fundamentally different tools for different problems.
Glue vs LinearB
LinearB measures engineering output. Glue gives code intelligence to the whole team — PMs, EMs, and devs.
Glue vs Cortex
Cortex builds service catalogs. Glue gives code intelligence to the entire product team. Different layers of the stack.
Glue vs Swimm
Swimm documents code. Glue gives intelligence across the whole product lifecycle. Here's the full breakdown.
Glue vs Jira
Jira tracks tickets. Glue shows you the code behind them. Why teams use both.
Glue vs Backstage
Backstage is a developer portal by Spotify. Glue is code intelligence for the whole product team. Different tools, different users.
Technical Debt Statistics 2026: The Data You Need to Convince Leadership
35% of CTOs name tech debt as their #1 challenge. Here's every stat you need to make the business case.
Code Dependency Analysis: Understanding Your System's Hidden Connections
Hidden dependencies cause 70% of estimation failures. Here's how to map them before they blow up your sprint.
Codebase Analysis Tools: A 2026 Buyer's Guide
The codebase analysis market is exploding. Here's what to look for and how the top tools compare.
Technical Debt Reduction: A Practical Playbook
You can't eliminate tech debt, but you can manage it. A step-by-step playbook for systematic debt reduction.
Code Health Metrics Every Engineering Manager Should Track
Your codebase has a health score — you just can't see it. Here are the metrics that matter and how to track them.
Technical Debt Tracking: Tools and Metrics That Work
You can't fix what you can't measure. Here's how to track technical debt with the right tools and metrics.
Bus Factor = 1: Stop Your Team's Knowledge Walking Out the Door
One developer leaving can tank an entire module. Here's how to measure and fix your bus factor before it's too late.