model evaluation

Adelphi delivers production AI solutions for national security, connecting models and agents to governed mission data across classified, covert and enterprise environments. The company provides a secure access layer and a governed registry of data sources, permissions, policies and provenance, and embeds cleared engineers to deploy AI into mission networks through accreditation and go-live.

accreditation engineeringagentic AI workflowsAI deploymentair-gapped deploymentdata access and integrationsecure pipelines and access controls

Bru™ was designed specifically for AI projects with focus on 3D medical images. It expedites two stages in the AI project lifecycle: data labeling, via a web viewer for labeling 3D medical images that employs algorithmic tricks including AI to increase labeling speed and provide convenient means for label overseeing; and no or low code deployment of a trained Neural Network, which can be as straightforward as uploading a file. Once uploaded, the NN becomes available for the Bru™ web viewer for evaluation by AI team members and for prediction through REST

AI-assisted labelingdata labeling of 3D medical imagesNN prediction through RESTful APIno or low code deployment of trained neural networkssupport for fused modalities such as PET-CT

ATOZ SERVICE, INC is a U.S.-based small business founded in 2021, focused on applied AI research, privacy-aware data systems, software prototyping, and technical validation. It develops Local LLM, RAG, data, and secure software prototypes for research validation, public benefit, and commercial transition, and is led by a principal with over 25 years of experience in software engineering, data systems, and technical program leadership.

AI & Local LLM R&DAWS/Azure cloud managementCybersecuritydata engineering and ETLManaged It Servicesnetwork managementNLP-based information retrievalRAG and vector searchsecure web and mobile developmentShopify/e-commerce developmentsoftware prototypingvoice AI

Centaur.ai provides high-quality, trusted data annotation at scale to power the development of superior AI models. Its methodology combines human judgment from a global network of subject matter experts with AI and quality measurement, drawing on research from the MIT Center for Collective Intelligence. The company was formerly known as Centaur Labs.

aggregation of human and AI annotationsdata annotationdata labelingdataset designhuman-in-the-loop QAmulti-modal annotation

ClydAI Technologies is a Virginia-based AI strategy and architecture firm that designs and builds AI systems. It helps federal, nonprofit, healthcare, and regulated enterprise teams identify practical AI opportunities, evaluate models, and design secure AI systems for cloud, on-premises, and air-gapped environments. The firm was founded by Khyati Nayak, who has more than 20 years of experience across software engineering, management consulting, and enterprise AI strategy.

AI governanceAI platform developmentAI strategyAI transformationEnterprise architecturesolution architecture

Defined.ai is an AI data marketplace and services provider that enables enterprises to buy or commission high-quality training data for artificial intelligence, including speech, text, image, video, and multimodal datasets. It offers custom data collection, data annotation, model evaluation, machine translation, and conversational AI solutions, with a focus on multilingual and domain-specific data. The company states it is ISO 27001, 27701 and 42001 accredited and GDPR and HIPAA compliant.

data annotationdata collectionhuman-in-the-loop workflowsLLM fine-tuningmachine translationmultimodal data collection

Digital Divide Data (DDD) delivers end-to-end ML data operations and training data solutions for large language models and computer vision systems, combining human expertise with advanced workflows. It serves Fortune 500 enterprises and global innovators in Physical AI, Generative AI, and Data Services, providing human-in-the-loop data annotation, validation, and evaluation.

3D LiDAR annotationdata collectiondata curationData engineeringdata orchestrationdata validationdata visualizationfine-tuninghuman preference optimizationimage annotationmultilingual NLPmultisensor fusionOCR and conversionsensor data annotationspeech/audio annotationtext annotationtranscriptiontranslationvideo annotation

Hewani Data delivers human-verified data for AI teams, providing end-to-end data annotation, validation, and ML/GenAI engineering. The work is done by a full-time, trained team hired and trained from East African talent, rather than crowdsourced workers. Services span annotation, QA, analytics, and ML system design, with project management run from San Jose, California and a delivery hub in Nairobi, Kenya.

3D point-cloud labelinganalytics & data sciencedata cleaning & structuringdata validation & QADevOps supportfractional CDAOGenAI system designimage annotationimage & instance segmentationML proofs of conceptprompt & response evaluationRed TeamingRLHF & preference rankingtext & NLP annotationvideo annotation

Hitherto AI develops ethical, relational, and enduring forms of intelligence for organizations and research environments. Its work spans research, architecture, consultancy, and education, including AI alignment, model evaluation, federal and enterprise training, and red teaming. It operates through Flux Hall (learning wing), Reason Forge (ethics lab and philosophical gymnasium), Friction Lab (trauma-informed red teaming), and Fathom Lab (closed R&D facility).

AI alignmentcomplex systems modelingcurriculum developmentedge inferenceelectromagnetic sensinginstrumentation designnon-destructive testingpolymer characterizationRed Teamingsensor fusionspectroscopysynthetic biologysynthetic dataset generationSystem architecturetraining and upskilling