SBA · SBA-34 · record R3190

Developer Code Assistant AI

Small Business Administration · OCIO: Office of the Chief Information Officer

Overview

Development stage: In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not high impact: This AI system does not meet the criteria of any of the six pillars that make up High Impact AI in Memorandum M-25-21.
Start date: 2026-01-01 (day precision)Source: 01/01/2026 00:00:00
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Information Technology
Operational functions: Software development assistance, Translation and transcription Derived by keyword rules; see methodology

Problem the AI is intended to solve

Developers spend substantial time writing repetitive boilerplate code, looking up syntax and APIs, translating requirements into code, and debugging errors. Understanding unfamiliar codebases is time-consuming. Manual code reviews are labor-intensive. Command-line operations require remembering complex syntax. Context-switching between documentation and coding disrupts flow.

Expected benefits

Increases developer productivity by 55-94% according to customer studies (Grupo Boticário case). Accelerates coding with real-time intelligent suggestions that understand context. Reduces time searching for documentation and syntax. Enables autonomous coding through GitHub issue assignment to AI agents. Provides instant code explanations and error detection. Improves code quality through AI-powered refactoring suggestions. Streamlines terminal workflows with natural language CLI commands. Maintains flow state by reducing context-switching. Supports multiple programming languages and frameworks.

System outputs

AI-powered pair programmer that provides context-aware code suggestions and completions directly in the code editor and terminal. Uses machine learning trained on billions of lines of public code to suggest entire lines, functions, and code blocks as developers type. Features autonomous coding agent that can be assigned GitHub issues to independently write code and create pull requests. Provides inline chat for code explanations, error detection, refactoring suggestions, and optimization recommendations. Supports natural language CLI commands. Analyzes context from open files, repository structure, and cursor position to generate probabilistic suggestions. Input: code context, natural language prompts, GitHub issues, repository files. Output: code suggestions, full functions, pull requests, refactoring recommendations, terminal commands, code explanations.

Technology

AI classification: Generative AISource: Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code).
System name(s): GitHub Copilot
Custom-developed code: No
Public source code: Not applicable - No custom code

Sourcing

How it was built: Purchased from vendorSource: a) Purchased from a vendor
Vendor (as reported): Microsoft Corporation (GitHub)
Vendors (standardized): MicrosoftGitHub

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: Yes
Privacy Impact Assessment: Not publicly available
Authorization to Operate: Yes
Demographic features: k) None of the above
Training and evaluation data: Commercial vendor-managed training data including public code repositories and technical documentation
Federal Data Catalog: Not applicable - Proprietary vendor training data

Governance

8 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingReported in placeSource: a) Yes

AI impact assessmentReported in placeSource: a) Yes

Independent reviewReported in placeSource: a) Yes – by another appropriate agency office or reviewer not directly involved in the AI’s development

Ongoing monitoringReported in placeSource: a) Yes, sufficient monitoring protocols have been established

Operator trainingReported in placeSource: a) Yes, sufficient and periodic training has been established

Fail-safeReported in placeSource: a) Yes

Appeal processNot applicableSource: b) Not applicable

User and public consultationOther or ambiguous answerSource: Conducted extensive pilot testing with software development teams, monitored code suggestion acceptance rates and developer productivity metrics, collected feedback through developer surveys and interviews, established security review processes for AI-generated code, incorporated feedback from code review findings, tracked autonomous agent performance on assigned issues, and continuously improved through analysis of developer interaction patterns and rejection reasons.Flag: free-text answer

Potential impacts and how they were identified

Potential impacts include code quality, introduction of security vulnerabilities or licensing issues from training on public code, and autonomous agent changes to repositories. Impacts identified through developer testing, code review processes, security analysis, and legal review of licensing implications. Mitigation includes mandatory peer review of AI-generated code, security scanning tools, license compatibility checking, developer training on code validation, branch protection rules, and human oversight of autonomous coding agent pull requests.