Software Developer
ARCHIVED
We can't find an active application page for this role right now. It may reopen or be listed elsewhere. Use Next Steps to search for an active apply link and similar live jobs.
Location Branson, MODepartment EngineeringOverviewWe’re seeking a skilled developer to work with the Whisper Automatic Speech Recognition (ASR) codebase and optimize it for real-time captioning, transcription, and translation applications. This role will focus on improving performance, latency, and accuracy.Key ResponsibilitiesWork directly with the Whisper ASR open-source model and underlying PyTorch codebase.Optimize inference speed and memory efficiency for real-time or near-real-time transcription.Implement low-latency streaming pipelines for audio capture, buffering, and incremental transcription.Profile and improve GPU/CPU utilization and model quantization or pruning where applicable.Develop and maintain API endpoints, CLI tools, or SDKs for Whisper-based real-time services.Work collaboratively with hardware, firmware, and product teams to ensure compatibility with on-premise embedded systems.Stay current on emerging speech recognition architectures and contribute ideas for R&D improvements.Conduct performance benchmarking, regression testing, and documentation of all optimizations.Required Skills & QualificationsStrong experience with Python and PyTorch (TensorRT).Familiarity with OpenAI Whisper or other ASR architectures (wav2vec2, Conformer, DeepSpeech, etc.).Experience with real-time or streaming data processing (e.g., WebSockets, asyncio, FFmpeg).Experience working with LLMs and designing robust systems that utilize them.Background in CUDA optimization and GPU acceleration techniques.Understanding of audio preprocessing, signal processing, and speech-to-text workflows.Experience working in Linux environments.Strong debugging, profiling, and performance tuning skills.Preferred QualificationsPrior work on low-latency broadcast or captioning systems.Knowledge of python and C++ for performance-critical components.Experience deploying models on embedded devices (Jetson, Orin, or similar).Familiarity with real-time translation or multi-language ASR.Location:In-office preferred.Hybrid or remote may be considered based on applicant.BenefitsMedical, dental, and vision insurance (70% company paid premium for employee)Short term disability and life insurance (100% company paid premium for employee)Retirement plan with 3% company match.PTO (based on length of employment) + 5 sick/personal days + 7 company holidays.How To ApplyPlease submit your resume, a cover letter explaining your interest in the project, and any relevant portfolio or GitHub links toDave Kendall atdkendall@linkelectronics.comand Dominic Armelie atdarmelie@linkelectronics.com.Company OverviewFounded in 1989, Link Electronics has established itself as a leader in developing quality, competitive products for the broadcast industry. With a history of innovation in audio and video applications, our product line has grown to over two hundred offerings, adapting to the latest technologies and industry standards. Serving a broad range of sectors, including television broadcasting, municipalities, educational and religious institutions, and closed captioning services, Link Electronics prides itself on exceptional customer service, quality products, and competitive pricing. As we continue to introduce new technologies, our commitment remains strong: to be an industry leader while delivering cost-effective solutions to our customers.