Civil Servants as Builders: Enabling Non-IT Staff to Develop Secure Python and R Tools

Prashant Sharma · 2025

Read Civil Servants as Builders: Enabling Non-IT Staff to Develop Secure Python and R Tools by Prashant Sharma (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07203.txt.

Introduction

Prashant Sharma describes a platform that lets government staff who can code, but sit outside IT, build their own small tools. Digital government research centers on pro IT teams, vendors, and no-code products. A middle group — skilled civil servants with no sanctioned path to deploy code — goes unstudied. His 2025 paper fills that gap with an open-source design.

The platform runs on Jupyter notebooks and preapproved libraries, inside a sandboxed and auditable workflow. Peer review is built in, and the setup respects procurement and security rules without locking the state to a vendor. Unlike no-code tools, it keeps staff skills sharp and growing. Public administrators get a replicable model for safe in-house building.

Catalog record

CREATORPrashant Sharma
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE41 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-07203

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Civil Servants as Builders: Enabling Non-IT Staff to Develop Secure Python and R Tools Prashant Sharma Abstract . Current digital government literature focuses on professional in-house IT teams, specialized digital service teams, vendor-developed systems, or proprietary low-code/no-code tools. Almost no scholarship addresses a growing middle ground: technically skilled civil servants outside formal IT roles who can write real code but lack a sanctioned, secure path to deploy their work. This paper introduces a limits-aware, open-source and replicable platform that enables such public servants to develop, peer-review, and deploy small-scale, domain-specific applications within government networks via a sandboxed, auditable workflow. By combining Jupyter Notebooks, preapproved open-source libraries, and lightweight governance, the platform works within institutional constraints such as procurement rules and IT security policies while avoiding vendor lock-in. Unlike low/no-code approaches, it preserves and enhances civil servants’ programming skills, keeping them technically competitive with their private-sector peers. This contribution fills a critical gap, offering a replicable model for public-sector skill retention, resilience, and bottom-up digital transformation. Conference: 11th Workshop on Computing within Limits; June 26–27, 2025; Online 1. Introduction Government agencies operate under important mandates: to safeguard public resources, ensure procedural fairness, and uphold high standards of security. To meet these obligations, digital tools in the public sector are typically subjected to rigorous procurement procedures and strict IT oversight. These institutional safeguards exist for good reason—they are designed to prevent waste, ensure equitable access to contracting opportunities, and reduce risks to sensitive data and infrastructure. However, they also have a cost: civil servants who identify clear opportunities for internal automation or lightweight tooling often face long procurement cycles or compete for limited IT bandwidth already stretched thin. Procurement departments, for instance, are responsible for ensuring that every expenditure serves the public interest and that projects do not bypass fair competition. Similarly, IT teams act as custodians of security and stability, ensuring that all deployed tools conform to internal standards and do not introduce cybersecurity vulnerabilities. In environments where time, staffing, and attention are chronically limited, both functions tend toward caution. This tendency, while understandable, can unintentionally suppress creativity, especially when civil servants propose novel solutions that fall outside conventional procurement or IT service delivery pipelines. Today, many civil servants possess technical skills, such as scripting in Python or R, querying internal databases, or creating dashboards using open source libraries. However, these skills often remain siloed within individuals or teams. When a civil servant builds a script that automates a complex task or provides interactive visualizations, there is often no sanctioned pathway to scale that solution. Formalizing even a simple application can require navigating approvals, documentation, security reviews, and infrastructure constraints that are disproportionate to the complexity of the tool. As a result, practical innovations built by domain experts often remain trapped in local files or are discarded when the staff member leaves. In addition, some government IT departments may interpret security controls, such as those outlined in NIST SP 800-53 ( National Institute of Standards and Technology, 2020 ) , as requiring strict oversight of programming environments, resulting in outright bans on tools like Jupyter Notebook, Python, or R for non-IT teams. This paper proposes a speculative but feasible system that enables civil servants to safely build and deploy internal tools within institutional limits. In doing so, it gestures toward a broader shift in government digital transformation, moving beyond reliance on external vendors or centralized digital teams. The proposed platform allows civil servants to create applications in Jupyter Notebooks and deploy them as secure, interactive web applications accessible exclusively to authenticated users on the government’s internal network, whether on-site or connecting remotely through a secure VPN gateway. Key features include a curated list of preapproved Python, R, or other relevant language libraries, automated sandbox execution, lightweight peer review, and internal-only web deployment. These mechanisms are intended not to replace institutional oversight, but to make it more scalable, allowing frontline staff to contribute low-risk digital tools while maintaining accountability and information security. Rather than presenting digital transformation as a process that must be outsourced to vendors or specialized IT teams, this paper argues for a shift in perspective: one in which civil servants already inside the system are empowered to create sustainable, domain-specific applications under constrained conditions. This shift reflects the broader values of the Computing within Limits community, working within institutional, infrastructural, and ecological boundaries while prioritizing resilience, adaptability, and the reuse of local knowledge. The sections that follow articulate the structural barriers to such an approach and how the proposed system works with, rather than around, those constraints. 1.1. Structural Barriers to Digital Innovation Despite growing technical fluency among civil servants, systemic and organizational barriers often prevent frontline innovation from taking root. These barriers are not primarily technological; they are deeply embedded in the structures of public administration, which tend to prioritize procedural integrity, risk minimization, and stability over agility and experimentation. Digital transformation in the public sector frequently proceeds through external procurement. While necessary for ensuring fairness and fiscal accountability, procurement processes are poorly suited to iterative or exploratory work. Edler et al. ( Edler et al., 2015 ) note that procurement regimes emphasize compliance and predictability rather than creativity or impact. This orientation often makes it difficult to contract for lightweight, open-ended tools or small-scale internal prototypes. Proposals that do not fit neatly into formal scopes of work or budget categories may be filtered out early, regardless of their practical utility. At the same time, government IT departments face pressures to maintain operational stability across critical infrastructure. Bozeman and Kingsley ( Bozeman and Kingsley, 1998 ) describe how public sector organizations develop strong cultures of risk aversion—where the political and reputational consequences of software failure far outweigh the rewards of incremental success. As a result, even technically feasible proposals involving new platforms, packages, or data access methods are often denied. This caution systematically excludes open-source or experimental tools, even when they are popular, well-documented, and widely used in academia or industry. Furthermore, digital innovation is not analogous to constructing a physical asset, such as a bridge or a building, that can be completed and then handed over for maintenance. Mergel ( Mergel, 2017 ) emphasizes that digital systems require continual refinement, adaptation, and support. When digital services are delivered entirely through third-party contracts, agencies often lack the institutional memory or technical fluency to modify them in response to shifting policy or operational needs. This dependency leads to brittle systems and long procurement delays for even minor updates—an unsustainable model under conditions of resource constraint. Legacy systems further intensify the challenge. As noted by the UK House of Commons Public Accounts Committee ( Walker, 2025 ; of Public Accounts, 2025 ) , much of government infrastructure relies on aging software that is difficult to modernize and expensive to maintain. These systems consume the bulk of IT budgets, leaving little room for experimentation or small-scale innovation. Moreover, their rigidity often requires new tools to conform to outdated standards, thereby limiting what frontline developers can feasibly build. These constraints are not temporary; they are structural. Yet as the Computing within Limits community has emphasized, working within limits does not mean abandoning innovation—it means redirecting it ( Tomlinson and Aubert, 2017 ;

Read the complete text