The rapid expansion of digital services has placed unprecedented pressure on data center design teams to deliver facilities that are both highly reliable and adaptable to shifting workloads. Traditionally, this process relied heavily on manual drafting cycles, repeated validation steps, and fragmented data sources that made scaling difficult and introduced opportunities for human error. As organizations pursue greater efficiency, the need to embed repeatable, automated practices into the core of engineering workflows has become a strategic imperative rather than a optional improvement. This shift is not merely about speeding up individual tasks; it is about redefining how knowledge is captured, shared, and applied across teams working on complex infrastructure projects.
Modern data center projects demand coordination among electrical, mechanical, and IT specialists, each contributing layers of detail that must align precisely. When these contributions remain siloed in individual expertise, inconsistencies creep in, and updating designs to reflect new standards or equipment specifications becomes a labor‑intensive ordeal. The industry has long recognized that manual processes create bottlenecks, yet many firms continue to rely on the tacit knowledge of senior designers, which is difficult to transfer and often undocumented. Addressing this gap requires a deliberate move toward treating design elements as data that can be programmed, validated, and reused without losing the nuance that experienced engineers bring to the table.
Janvi Saddi exemplifies the engineer‑developer hybrid who bridges domain expertise with software engineering to tackle these systemic challenges. Her background spans CAD automation, structured data handling, market research, and AI‑assisted productivity, giving her a unique vantage point on where manual effort creates friction. Rather than accepting the status quo, she began by mapping repetitive design activities—such as generating tray polylines from CSV inputs, rendering labels, converting blocks, and verifying conduit lengths—to identify where automation could replace error‑prone manual steps while preserving design intent.
She built a suite of more than a dozen custom AutoCAD commands using a blend of AutoCAD API, .NET, AutoLISP, C#, and DLL plug‑ins, each targeting a specific recurring task. By encapsulating these operations within programmable commands, she turned what used to be a series of mouse clicks and manual checks into repeatable scripts that can be invoked with a single keystroke. This approach not only reduced the time spent on routine drafting but also introduced a layer of consistency that is difficult to achieve when each designer interprets specifications slightly differently.
A cornerstone of her solution was the adoption of JSON‑based structures to standardize design parameters across projects. By storing key attributes—such as tray dimensions, block properties, and validation thresholds—in a universal, human‑readable format, the automation logic could read, modify, and reuse these parameters without being tightly coupled to any single drawing file. This decoupling made it easier to propagate updates across multiple designs, ensured that all teams worked from the same baseline, and significantly reduced the risk of version drift when specifications evolved.
To encourage adoption, Janvi designed a custom AutoCAD ribbon that surfaced her automation tools in a familiar, accessible interface. She accompanied the rollout with pilot testing sessions, gathered feedback from practicing designers, and refined the workflows based on real‑world usage. This user‑centric approach helped overcome the natural resistance to new tools, as designers could see immediate benefits in their daily work while still operating within the AutoCAD environment they knew well.
The impact of these changes is quantifiable and striking. Block adaptation, which previously consumed nearly a full day in recent workflows and stretched to almost two months in legacy processes, now completes in under five minutes. Likewise, exporting critical CAD models—a step that once took roughly eight minutes—has been accelerated to approximately 300 milliseconds, representing a 96% reduction in time. These gains are not merely incremental; they fundamentally alter project timelines, allowing teams to iterate designs faster, respond to change requests with agility, and allocate more effort to higher‑value activities such as performance optimization and risk assessment.
Beyond raw speed, the integration of validation directly into the design process has improved accuracy. By embedding checks for conduit length, attribute completeness, and geometric consistency within the automation scripts, potential issues are flagged at the moment of creation rather than discovered during later review cycles. This early detection reduces costly rework, minimizes the likelihood of field‑installation surprises, and fosters a culture where quality is built into the workflow rather than inspected after the fact.
Janvi’s method also addresses the perennial challenge of knowledge capture in engineering organizations. Instead of relying on individual designers to remember specific procedures, her workflow converts expert knowledge into structured data and reusable logic. JSON parameter files, automated block conversions, and geometry‑based checks serve as executable documentation that can be version‑controlled, audited, and shared across teams. This transformation makes the organization less dependent on the presence of any single expert and more resilient to turnover or shifting project demands.
Artificial intelligence has played a supportive role in accelerating the development of these automation tools. Janvi leveraged AI‑assisted coding assistants such as Gemini and GitHub Copilot to generate boilerplate code, debug scripts, and produce documentation more rapidly. Importantly, she positioned AI as a augmentative force—providing suggestions that are then subjected to engineering judgment, rigorous testing, and validation. This balanced approach ensures that the resulting tools are both innovative and reliable, avoiding the pitfalls of over‑reliance on opaque AI outputs.
The lessons from her work extend well beyond data center design. In an earlier market research role, Janvi tackled similar inefficiencies by standardizing multilingual survey processes, introducing A/B testing, and building a centralized data warehouse with Python‑based conversion scripts that saved roughly 40 hours per month. Though the domains differed, the underlying pattern—repetitive, fragmented processes crying out for systematization—remained identical. This cross‑domain experience reinforces the idea that automation principles are transferable and that investing in standardized data models yields dividends across various technical functions.
Looking ahead, engineering firms that manage increasingly complex infrastructure and swelling volumes of technical data will find that the combination of domain expertise, programming skill, structured data, validation logic, and judicious AI assistance becomes a core competency. Organizations seeking to replicate this success should start by mapping their most repetitive, error‑prone tasks, invest in building reusable automation libraries, adopt lightweight data standards like JSON, and foster a culture where continuous feedback refines the tools. Pilot projects with clear metrics—such as time saved, error reduction, and user adoption—provide the evidence needed to scale automation initiatives enterprise‑wide.
Actionable advice for leaders: begin with a workflow audit to pinpoint high‑frequency manual steps; create a small, cross‑functional team of designers and developers to prototype automation for one of those steps; define measurable success criteria before implementation; use version‑controlled repositories for scripts and parameter files; integrate validation checks directly into the automation to catch issues early; provide hands‑on training and accessible UI elements (such as custom ribbons) to encourage adoption; review results regularly and iterate based on user feedback; and finally, institutionalize the approach by embedding automation considerations into project planning and performance reviews. By following these steps, firms can transform their engineering workflows from brittle, expert‑dependent processes into scalable, measurable systems that keep pace with the demands of modern digital infrastructure.