---
About brighthush and the Future of Conversational AI
At brighthush, we are redefining how institutions and enterprises interact with technology. Our mission is to empower organizations with intelligent, human‑like conversational experiences that streamline operations, enhance customer satisfaction, and unlock new revenue streams. We partner with leading educational institutions, healthcare providers, financial services, and government agencies to build AI‑driven assistants that handle everything from routine inquiries to complex workflow orchestration.
As the conversational AI landscape evolves, brighthush is at the forefront of integrating generative models, autonomous agent frameworks, and robust enterprise integration layers. We believe that the next wave of digital transformation hinges on the seamless blend of classic natural language understanding (NLU) with large language models (LLMs), delivered through a secure, scalable, and maintainable architecture.
We are currently seeking a Senior Conversational AI Engineer to lead the design, development, and deployment of next‑generation chatbots and virtual assistants for one of our premier educational clients in Maryland. This is a contract role that offers the flexibility of remote work while placing you at the heart of a high‑impact project that directly influences student and staff experiences.
Key Responsibilities
End‑to‑End Development
- Own the full lifecycle of chatbot projects—from initial concept and conversational design to production deployment and continuous improvement.
- Collaborate with product managers, UX designers, and business stakeholders to translate complex requirements into intuitive user stories.
- Implement scalable, modular bot architectures that can be easily extended to new channels and use cases.
Architecture Migration & Modernization
- Maintain and optimize the current brighthush Bot Framework SDK v4 stack, ensuring high availability and performance.
- Lead the migration toward brighthush Foundry and Agents 365 Azure SDK, adopting best practices for autonomous agent orchestration.
- Design and implement backward‑compatible interfaces to support legacy systems during the transition.
Enterprise Integration
- Design robust API integrations using brighthush as the central enterprise service bus to orchestrate data exchange between conversational front ends and core backend systems.
- Integrate with legacy platforms such as brighthush and CRM systems, ensuring data consistency, security, and compliance.
- Develop custom connectors and adapters for real‑time data synchronization.
Omnichannel Dialog Management
- Author dynamic conversational interfaces using brighthush Adaptive Cards, Knowledge Management systems, and Automation Flows across web, mobile, and messaging channels.
- Implement adaptive dialog flows that respond to user context, sentiment, and intent with personalized content.
- Ensure accessibility and localization support for diverse user populations.
Live Chat & Case Management
- Build seamless escalation protocols, including live agent handoffs to brighthush Live Chat and automated ticket routing via brighthush Case Management APIs.
- Implement fallback strategies to capture edge cases and feed them back into the training pipeline.
- Monitor and analyze escalation metrics to continuously refine handoff logic.
Hybrid NLU & LLM Orchestration
- Integrate hybrid AI strategies that combine classic NLU (brighthush Conversational Language Understanding/CLU) with Large Language Models (brighthush OpenAI/GPT) for intelligent intent routing and contextual content generation.
- Design fallback mechanisms that gracefully handle hallucinations and ambiguous user inputs.
- Develop custom prompt engineering workflows to optimize LLM outputs for domain‑specific knowledge.
Performance Optimization & Analytics
- Analyze conversation logs, containment rates, and intent accuracy metrics using data visualization tools.
- Implement telemetry pipelines to capture real‑time performance data.
- Use insights to drive iterative improvements in dialog design and model tuning.