Highlight Technologies, Inc.

DOCUMENT FOR PUBLIC RELEASE

The decision issued on the date below was subject to a GAO Protective Order. This redacted version has been approved for public release.

Decision

Matter of: Highlight Technologies, Inc.

File: B-423837.3

Date: February 19, 2026

Devon E. Hewitt, Esq., and Matthew L. Nicholson, Esq., Potomac Law Group, PLLC, for the protester.

David S. Black, Esq., and Gregory R. Hallmark, Esq., Holland & Knight LLP, for Steampunk, Inc.; and Aron C. Beezley, Esq., Patrick R. Quigley, Esq., and Elizabeth A. Brown, Esq., Bradley Arant Boult Cummings LLP, for IntelliBridge, LLC, the intervenors.

Richard W. Postma, Esq., Jessica A. Easton, Esq., and Laura L. Chapman, Esq., Department of Homeland Security, for the agency.

Sarah T. Zaffina, Esq., and Heather Weiner, Esq., Office of the General Counsel, GAO, participated in the preparation of the decision.

DIGEST

Protest challenging the agency's evaluation of the protester's quotation under the solicitation's live demonstration coding challenge is denied where the evaluation was reasonable and consistent with the solicitation criteria.

DECISION

Highlight Technologies, Inc. (Highlight), a small business of Fairfax, Virginia, protests the issuance of task orders to IntelliBridge, LLC, a small business of McLean, Virginia, and Steampunk, Inc., a small business, of McLean, Virginia, under request for quotations (RFQ) No. 70SBUR24Q00000105, issued by the Department of Homeland Security (DHS), United States Citizenship and Immigration Services (USCIS), for development services to enhance and modernize the intake domain for the myUSCIS program. Highlight argues that the agency improperly evaluated its technical quotation under the coding challenge factor of the evaluation criteria, and therefore, the agency's best-value determination was unreasonable.

We deny the protest.

BACKGROUND

On September 9, 2024, pursuant to the procedures of Federal Acquisition Regulation subpart 8.4, the agency issued the RFQ to holders of the General Services Administration's federal supply schedule contracts for information technology (IT) professional services under special item number 54151S.[1] Agency Report (AR), Exh. 14, RFQ amend. 3 at 1‑2; Contracting Officer's Statement (COS) at 1. The RFQ contemplated the issuance of no more than two task orders for a 1‑year base period and four 1-year option periods, to provide support services in developing new capabilities and modernizing existing capabilities for DHS's IT systems and applications across the intake domain.[2] RFQ at 3; AR, Exh. 5, RFQ attach. 1, Performance Work Statement (PWS) at 5. The intake domain covers several related programs, including myUSCIS, concerned with form submissions, fee payments, and capturing required documents and information, such as “.pdf upload, third party software/application integration efforts, and other initiatives from both internal and external stakeholders.”[3] PWS at 5.

The RFQ provided for a multi-phase procurement. In phase 1, firms were to submit written quotations containing relevant business information and a corporate experience statement. RFQ at 49-52. The agency was to evaluate the phase 1 submissions before conducting an “advisory down‑select.” Id. at 52. The RFQ stated that the advisory down-select was only a recommendation and vendors that were advised not to proceed could chose to continue to phase 2. Id.

For phase 2, vendors were to participate in a coding challenge and submit a price quotation. Id. at 53-57. The coding challenge included a “take home” portion and a virtual, live demonstration. Id. at 52-53. The RFQ required each firm to create a web application solution to a problem statement and then conduct a live demonstration of the solution by deploying the code. Id. at 53‑54. During the live demonstration, the agency would also ask each firm “to implement enhancements or changes” to their solution and the firm would have three hours in which to implement and deploy the new requirements in the live demonstration. Id. at 53, 55.

The RFQ provided for award to the firms that gave the agency the “greatest confidence that [they] will best meet the requirements affordably in a way that will be most advantageous to the Government.” Id. at 57. The agency would use a best-value tradeoff analysis, considering the following three factors: (1) corporate experience; (2) code challenge; and (3) price. Id. The corporate experience factor was more important than the code challenge factor and when combined, the non-price factors were significantly more important than price. Id.

On October 9, 2024, the closing date for receipt of phase 1 quotations, Highlight timely submitted its quotation. COS at 2. DHS assigned a rating of high confidence to Highlight's quotation and extended an invitation to Highlight to participate in phase 2 of the competition. Id.

Nine firms, including Highlight, IntelliBridge, and Steampunk, participated in phase 2 of the competition. COS at 4. The live demonstration of Highlight's solution occurred on February 24. AR, Exh. 63, Highlight Demonstration Teams Meeting Transcript Part 1; AR, Exh. 64, Highlight Demonstration Teams Meeting Transcript Part 2.

The contracting officer, who was also the source selection authority (SSA), reviewed the evaluation findings and determined that IntelliBridge and Steampunk proved the best value overall. AR, Exh. 54, Source Selection Decision Document (SSDD) at 2, 6; COS at 4. On August 8, DHS notified Highlight that its quotation was not selected for the issuance of a task order. AR, Exh. 55, Unsuccessful Notice at 1.

Highlight filed a timely protest with our Office arguing that the agency's evaluation and best‑value tradeoff analysis were unreasonable. Highlight Techs., Inc., B‑423837.2, Aug. 29, 2025 (unpublished decision). We dismissed the protest because DHS notified our Office of its intent to take corrective action that would result in the issuance of a new award decision, and thus, rendered the challenges to the initial evaluation academic. Id. In this connection, DHS advised that it intended “to take a second look at the quotations, the evaluation and the notifications, and make a new source selection decision if necessary.”[4] AR, Exh. 56, Notice of Corrective Action at 1.

For the agency's corrective action, it issued stop work orders to IntelliBridge and Steampunk, and reevaluated quotations. AR, Exh. 60, Stop Work Order IntelliBridge; AR, Exh. 61, Stop Work Order Steampunk. The technical evaluation committee (TEC) reviewed the initial evaluation and the SSA prepared a revised SSDD. COS at 2, 4; Memorandum of Law (MOL) at 3. The reevaluation did not impact Highlight's evaluation. COS at 4. The final evaluation ratings for the quotations of Highlight, IntelliBridge, and Steampunk are as follows:[5]

Evaluation Factor

Highlight

IntelliBridge

Steampunk

Corporate Experience

High Confidence

High Confidence

High Confidence

Code Challenge

Some Confidence

High Confidence

High Confidence

Total Evaluated Price[6]

$237,865,427

$256,654,958

$259,100,334

AR, Exh. 54, SSDD at 5. The SSA concluded that the awardees' superior non-price quotations justified paying a higher price and represented the best value to the government. See COS at 7.

On October 29, DHS notified vendors that it had reissued the task orders to IntelliBridge and Steampunk following the reevaluation. Id. at 5. After receiving an expanded explanation of the evaluation from DHS on November 7, Highlight filed this protest with our Office. Id.; AR, Exh. 59, Post‑Corrective Action Expanded Explanation.

DISCUSSION

Highlight challenges the agency's evaluation of its technical quotation. The protester argues that DHS erred in assigning its quotation a rating of some confidence under the code challenge factor. Protest at 8‑11. Highlight also argues that the agency failed to consider aspects of its quotation that exceeded solicitation requirements and that its quotation merited a rating of high confidence. Id. at 11. Highlight contends further that the agency's best-value tradeoff analysis was unreasonable because it was based on the unreasonable rating of some confidence assigned to its quotation under the code challenge. Id. at 12. Based on our review of the record, we find no basis to sustain the protest. While we do not address each of the protester's allegations and variations thereof, we have reviewed them all and find the agency's evaluation reasonable and consistent with the solicitation's evaluation criteria.

The evaluation of quotations is a matter within the discretion of the procuring agency. eKuber Ventures, Inc., B‑420877, B‑420877.2, Oct. 13, 2022, at 4. The agency is responsible for defining its needs and the best method of accommodating them. Digital Sys. Grp., Inc., B-286931, Mar. 7, 2001, at 7. Our Office does not independently evaluate quotations; rather, we review the agency's evaluation to ensure that it is consistent with the terms of the solicitation and applicable statutes and regulations. eKuber Ventures, supra. A protester's disagreement with the agency's judgment, by itself, is not sufficient to establish that an agency acted unreasonably. Id.

Challenge to Adjectival Rating Under Code Challenge Factor

Highlight challenges the agency's evaluation of its quotation under the code challenge factor. In particular, Highlight disputes the agency's finding that its solution failed to deploy to the production environment during the live demonstration and warranted a rating of some confidence. Protest at 9-11. DHS responds that its evaluation of the protester's solution was reasonable and based upon the stated evaluation criteria. COS at 5-6; MOL at 4‑21; AR, Exh. 62, Decl. of TEC Chairperson. As discussed below, we find the agency's evaluation unobjectionable.[7]

As described above, the RFQ provided that vendors would participate in a code challenge for phase 2 of the competition that included a “take home” portion and a virtual, live demonstration. RFQ at 52. Firms had 14 days to create a solution “capable of receiving USCIS [form] I-131 data via web based data entry as well as the ability for scanned evidence (.img, .jpeg, .pdf, etc.) to be uploaded from the USCIS customers.”[8] Comments exh. A, Code Challenge Instructions at 1. The vendor's solution must be able to stream data available for USCIS or interfaces that enable “data for integration.” Id. The “solution shall also include a chat-bot service to assist with the USCIS customers' experience.” Id. The solicitation instructed vendors to store the solution's source code in a GitHub repository and deploy the application in the vendor's Amazon Web Services (AWS) cloud environment.[9] Id.; RFQ at 54-55. The solicitation further advised that during the live demonstration, DHS would provide additional instructions and the vendor would be given three hours to implement the changes and then demonstrate its solution. RFQ at 55.

The RFQ provided that the vendor's proposed solution must demonstrate its ability to successfully design, program, test, and deploy the assigned task. RFQ at 53. The code challenge consisted of five evaluation criteria: (1) architectural and technical approach, (2) DevSecOps,[10] (3) user‑centered design and engineering, (4) logic and code quality, and (5) teaming and processes. Id. at 53, 59. The RFQ also identified additional evaluation details related to the evaluation criteria, including as relevant here, DevOps & continuous integration/continuous deployment (CI/CD),[11] and test code coverage.[12] Id. at 53, 59‑61. In this context, the solicitation provided that DHS would evaluate how well the firm “followed instructions, and whether it successfully architected, built, and deployed an open-source solution.” Id. at 53, 59-61.

As relevant here, with regard to DevOps & CI/CD, the solicitation required that vendors “[i]mplement DevSecOps and solve the solution using a full-stack cloud‑native 100 [percent] automated CI/CD pipeline with a minimum of three (3) environments (development, stage, and production) and gates (e.g., automated tests must pass in lower environments, or the code will not be promoted to the next environment). Id. at 60. With regard to test code coverage, the solicitation required that vendors “produce[ ] automated tests for written software that verify desired functionality” and that the “[p]ercentage of edits and validations in the code is covered by test cases.” Id. Notably, the solicitation did not require a set test code coverage threshold; rather, it allowed vendors to select their own coverage thresholds to “verify desired functionality.” Id.; AR, Exh. 62, Decl. of TEC Chairperson at 5 (explaining that firms “set their own ‘gates' (which Highlight set at 90 [percent] in this case)). The solicitation then required vendors to meet that threshold. See RFQ at 60; AR, Exh. 62, Decl. of TEC Chairperson at 5.

The RFQ provided that during the live demonstration of the vendor's solution, DHS would “assess its level of confidence in the [vendor's] ability to create and deliver business value utilizing DevSecOps practices, collaborate within the team, meet business and technical requirements, create innovative approaches and practices, and deliver products efficiently based on the [vendor's] working application.”[13] RFQ at 61.

During the live demonstration, Highlight's solution failed the automated tests and did not deploy changes to the production environment as required by the solicitation. AR, Exh. 50, TEC Report at 13‑14. Specifically, the protester's code for a new chat box capability did not pass the 90 percent test code coverage threshold, i.e., it failed the gate, and therefore, was not automatically promoted to the production environment. Id. at 13; Comments at 8 n.9. Instead, Highlight was forced to manually bypass the failed code gate to deploy the solution. AR, Exh. 50, TEC Report at 13‑14. In this context, Highlight was able to demonstrate the chat box function in a lower environment--the staging environment--but in order to demonstrate the solution in the production environment as required by the RFQ, Highlight had to bypass the failed gate. Id. at 13‑14; Comments at 8 n.9.

In evaluating Highlight under the code challenge, the agency explained that bypassing a failed gate to deploy a solution in the production environment potentially allows bugs in the production environment because code is promoted to production without meeting code quality standards. See AR, Exh. 50, TEC Report at 14. In this regard, the agency noted that Highlight's failure to automatically deploy its solution in the production environment occurred because Highlight set the test code coverage at 90 percent, a high coding standard, and its coding build did not meet this threshold. Id. at 13‑14.

Ultimately, the evaluators assessed a rating of some confidence to Highlight under the code challenge factor. Id. at 13. The evaluators observed that Highlight's solution met the baseline requirements for the five evaluation criteria but that its submission provided “only a few significant elements to inspire overall increased confidence.” Id. As Highlight selected a 90 percent test code coverage, the evaluators noted that Highlight's failure to deploy its solution indicated “challenges of maintaining high code quality standards and ensuring that all necessary tests pass before deployment.” Id. The evaluators also noted that failing the code testing quality “underscores the importance of thorough testing processes and validation gates to prevent incomplete or unverified changes from impacting production systems.” Id. The evaluators concluded that Highlight's failure to deploy its solution validated their assessment of Highlight's solution and the government's rating of some confidence. Id. at 13-14.

The protester argues that its solution was deployed in the development, staging, and production environments, and asserts that “[t]he fact that an error occurred upon deployment in the production environment does not mean that the web application failed to deploy.” Protest at 10. Highlight acknowledges that it had to “‘bypass a failed quality gate' to deploy the new code to the production environment.” Comments at 8. In Highlight's view, however, the failure of its code to deploy in the production environment was the result of its enhanced security protections and its high test code coverage of 90 percent, which, the protester's asserts, would benefit the government and should not have resulted in a rating of some confidence. Protest at 10-11; Comments at 7-11.

Based upon our review of the record, we find nothing unreasonable regarding the agency's evaluation. The solicitation clearly set out the evaluation criteria and the requirements for the live demonstration, which included successful deployment of changes to the production requirement. RFQ at 55. The solicitation also explicitly required changes to be deployed progressively through the development, staging, and production environments by gates such that code failing to meet the code coverage test percentage would not be automatically promoted to the next environment. Id. at 60. During the live demonstration, Highlight was unable to deploy code implementing the new requirements without bypassing a failed gate. In this context, Highlight was only able to partially meet the RFQ requirements.

The contemporaneous record shows that the evaluators noted that Highlight had to bypass the failed gate to deploy its changes to the production environment and that its demonstration of the new requirement took place in the development requirement. AR, Exh. 50, TEC Report at 13-14; see also AR, Exh. 62, Decl. of TEC Chairperson at 6‑7 (explaining how Highlight demonstrated its chat box code changes in the production and staging environments and referring to the Teams Meeting Transcript); AR, Exh. 64, Highlight Demonstration Teams Meeting Transcript Part 2 at 91. The evaluators acknowledged that the failure to deploy the solution was based on the high test code quality threshold selected by the protester and the failure demonstrated the challenges associated with high code quality standards and ensuring all required tests pass prior to deployment. AR, Exh. 50, TEC Report at 13. According to the evaluators, Highlight's failure to deploy the changes and its ability to bypass the gate also emphasized how important testing and validation gates are to stop “incomplete or unverified changes” to the production environment. Id. at 13‑14. The evaluators concluded that Highlight's solutions did not “inspire overall increased confidence.” Id. at 13.

Although the protester attempts to minimize its failure to meet the solicitation requirements--arguing that “[t]he fact that an error occurred upon deployment in the production environment does not mean that the web application failed to deploy”; rather the error was the result of its security enhancements and a high test code coverage threshold--the protester has failed to demonstrate that its solution met the RFQ's requirements or that the agency's evaluation was unreasonable. Protest at 10.

As noted above, vendors set their own test code coverage thresholds and the solicitation required them to meet that threshold. See RFQ at 60; AR, Exh. 62, Decl. of TEC Chairperson at 5. Highlight selected a high test code threshold for its solution, but it could not meet the threshold in the demonstration for the new requirement. Further, Highlight's security enhancements resulted in its failure to deploy the chat box solution. Although the protester disagrees with the agency's evaluation of some confidence to its code challenge, the record reflects that Highlight's solution failed to deploy as required by the RFQ. The protester's disagreement with the agency's evaluation, without more, fails to establish that the evaluation was unreasonable. eKuber Ventures, supra. Accordingly, this protest ground is denied.

Challenge That Proposed Features Exceeded RFQ Requirements

Highlight argues that the agency's evaluation failed to properly consider features of its solution that, in the protester's view, exceeded the solicitation requirements.[14] Protest at 11. In this regard, the protester contends that its 90 percent test code quality threshold was above the 60 percent industry standard, and as such, should have been evaluated as exceeding the RFQ's requirements. Id. DHS argues that the solicitation did not set a 60 percent test code requirement and that there is no established industry standard. MOL at 22. The agency argues, moreover, that Highlight did not even meet the 90 percent test code coverage it set for itself. Id. at 2324. We find the agency's evaluation unobjectionable.

As discussed above, the solicitation did not require a set test code coverage threshold; rather, it allowed vendors to select their own automated tests to “verify desired functionality.” RFQ at 60; AR, Exh. 62, Decl. of TEC Chairperson at 5 (explaining that firms “set their own ‘gates' (which Highlight set at 90 [percent] in this case), but per the clear terms of the solicitation they had to meet their own self-imposed thresholds in order to pass in lower environments.”). As also discussed previously, Highlight set a high standard for itself but was unable to meet its standard, and as such, its solution failed to deploy in the production environment as required by the solicitation. The record shows that the evaluators recognized the positive aspect of the high test quality threshold because it “prevent[s] incomplete or unverified changes from impacting production systems.” AR, Exh. 50, TEC Report at 13. The record also reflects that the evaluators noted that Highlight's coding did not meet its high threshold. Id. In this regard, the evaluators found that Highlight's new code for the change did not deploy to the production environment and that the failure revealed the “challenges of maintaining high code quality standards.” Id.

Based on our review of the record, Highlight has not shown that the agency's evaluation was unreasonable. As an initial point, it is not clear that there is an industry standard for test code coverage. Compare Protest at 11 (alleging an industry standard of 60 [percent]), with Comments at 8-9 n.10 (citing sources recommending 80 [percent] coverage), and MOL at 22 (citing source that states “there is no ‘ideal code coverage number'”). It is clear, however, that the solicitation did not establish a test code coverage requirement and instead allowed vendors to create their own standard. Highlight was unable to meet its self-selected standard of 90 percent during the live demonstration and the TEC report includes both positive and negative findings about Highlight's 90 percent test code threshold. To the extent Highlight asserts that the agency should have evaluated its 90 percent test code threshold as providing a benefit to the government, the protester's disagreement with the agency's judgment does not, without more, show that the agency's evaluation was unreasonable. Accordingly, this protest ground is denied.

Best-Value Tradeoff Analysis

Highlight also argues that the agency's best-value tradeoff decision was improper because it was based on alleged flaws in the underlying evaluation of Highlight's live demonstration. Protest at 11. This allegation is based on the protester's various challenges to the agency's evaluation of its solution under the code challenge factor. As discussed above, we find no basis to object to the agency's evaluation. ITegrity Inc., B-422694, B-422694.2, Sept. 26, 2024, at 8 n.2 (“Where other challenges to an evaluation of quotations have been denied or otherwise dismissed, a derivative challenge to the best-value determination does not afford a basis to sustain the protest.”). Accordingly, we dismiss this allegation because it does not establish a basis to question the selection decision.

The protest is denied.

Edda Emmanuelli Perez

General Counsel

[1] The solicitation was amended three times; citations to the RFQ in this decision refer to the revised version of the solicitation issued with amendment 0003 unless otherwise noted. In addition, citations to the record use the Adobe PDF pagination of documents produced in the agency report.

[2] While the solicitation is identified as an RFQ, the parties and various documents in the agency report use the terms quotation and proposal interchangeably. The distinction between the terms has no bearing on our analysis of the issues presented; our decision refers to the submission of quotations by vendors.

[3] “myUSCIS is a public facing responsive website that provides a ‘one stop shop' for users to navigate their immigration journey.” PWS at 5.

[4] Another unsuccessful firm also filed a timely protest arguing that the agency's evaluation was unreasonable and failed to assess the impact of a corporate transaction that occurred during the task order competition. This protest was also dismissed consistent with DHS's corrective action because the commitment to reevaluating quotations and making a new selection decision rendered the protest of the initial evaluation academic. Customer Value Partners, Inc., B‑423837, Aug. 29, 2025 (unpublished decision quoting Sun Chem. Corp., B‑288466 et al., Oct. 17, 2001, at 12).

[5] The solicitation provided for the evaluation of the non-price factors as high confidence, some confidence, or low confidence. RFQ at 62.

[6] This is the total evaluated price before discounts from the apparent successful awardees. RFQ at 49 (allowing the agency to request price reductions at any time during the competition). After discounts, IntelliBridge's price was $225,622,612 and Steampunk's price was $214,991,945. AR, Exh. 55, Unsuccessful Notice at 2.

[7] Highlight also challenges the agency's finding that Highlight's exclusion of a directory from its quality assurance testing decreased confidence in Highlight's ability to successfully perform. Protest at 8-9. We have reviewed the record and the agency's response and find no basis to sustain this protest ground.

[8] Form I-131 is a USCIS form titled, “Application for Travel Documents, Parole Documents, and Arrival/Departure Records.” Individuals without United States citizenship, such as refugees or individuals with other temporary protected statuses, use this form to obtain permission for professional or personal travel outside the United States. This form is also used by individuals who may be inadmissible or otherwise ineligible for admission into the United States, i.e. seeking parole, to gain admission to the United States for urgent humanitarian reasons or significant public benefit such as medical care or legal proceedings. I-131, Application for Travel Documents, Parole Documents, and Arrival/Departure Records, USCIS, available at https://www.uscis.gov/i-131 (last visited Feb. 17, 2026).

[9] GitHub is a web-based version control and collaboration platform for software developers that stores source code and tracks change history. See NCI Info. Sys., B‑416926 et al., Jan 9, 2019, at 2 n.2.

[10] “‘DevSecOps stands for development, security, and operations. It is an extension of the DevOps practice. Each term defines different roles and responsibilities of software teams when they are building software applications.'” IntelliBridge Comments at 2 n.4 (quoting What is DevSecOps?, AWS, available at https://aws.amazon.com/what-is/devsecops/ (last visited Feb. 19, 2026)).

[11] “‘Continuous integration automates the way developers consolidate changes in a shared code repository,' while ‘[c]ontinuous delivery ensures that the code validated at the continuous integration phase is ready for production deployment.'” IntelliBridge Comments at 2 n.5 (quoting What is CI/CD?, AWS, available at https://aws.amazon.com/what-is/ci-cd/ (last visited Feb. 19, 2026)).

[12] Specifically, the RFQ identified the following 12 evaluation details related to the evaluation criteria: code quality, accuracy, reusability, performance, scalability, DevOps & CI/CD, architectural considerations, design considerations, test code coverage, reporting, security, and product management. RFQ at 53, 59-61.

[13] The solicitation defined the rating of “high confidence” as follows: The Government has High Confidence that the [vendor] understands the requirement, proposes a sound approach, and will be successful in performing the task order with little or no Government intervention. RFQ at 62. For a rating of “some confidence,” the definition was the same except that the government has some confidence in the vendor and some government intervention is needed for the vendor's successful performance. Id.

[14] Although the protester argues that its quotation included multiple features that exceeded the solicitation requirements, the protester only discusses one feature, the high test code coverage threshold. Protest at 11.