Machine Learning Signal Processing Engineer at DeepSig Inc

Arlington, Virginia, United States -

Full Time

Start Date

Immediate

Expiry Date

04 Jan, 26

Salary

0.0

Posted On

07 Oct, 25

Experience

2 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Machine Learning, Wireless Signal Processing, DSP, 3GPP Standards, AI-RAN, Deep Learning, Python, C++, RF Sensing, Statistical Signal Processing, MIMO Systems, Beamforming, Model Compression, Integration, Collaboration

Industry

Software Development

Description

Description DeepSig Inc. is a venture-backed technology company pioneering the use of AI in wireless applications (physical layer communications, sensing, and others) by replacing traditional signal processing with algorithms derived by machine-learning (ML). DeepSig software products are achieving significant performance increases while reducing power consumption, which brings significant value to our customers. We are seeking engineers with strong expertise at the intersection of machine learning, wireless signal processing/DSP, and 3GPP standards. This role will support AI-RAN component development for next-generation RAN systems, including development of AI-RAN reference designs, optimized algorithms and implementations, dataset and training pipelines, input into 6G study items, and accelerated compute functions. You will work on both fundamental ML-for-PHY/AI-COMMS problems and software integration into RAN stacks, contributing to both open and commercial components, and interoperable, and high-impact next generation wireless systems. Key Responsibilities: AI-RAN Algorithm and Software-Module Development o Design and implement AI/ML models for multiple key RAN functions including receivers, spectrum sensing, ISAC/active sensing, schedule and resource optimization, and other use cases. o Contribute to algorithm optimizations for MU-MIMO/mMIMO optimization functions (e.g. user-pairing, prediction, beamforming, etc.). o Build reference ML module interfaces (data capture, online inference, scoring/benchmarking, simulation, performance validation) to integrate into RAN software stacks. Dataset & Training Infrastructure o Develop measurement and dataset collection tools and pipelines for training and scoring AI-RAN models and performance. o Build model training and KPI benchmarking tools for reproducible comparison across models and use cases and interoperable model testing. o Lead and contribute to commercial and open-source software for NextG AI-RAN capabilities. Accelerated Compute Functions o Implement and optimize critical baseband and AI/ML algorithms on accelerated compute platforms (e.g. GPU, NPU, TPU) with an emphasis on real-time deployment, latency, energy. o Explore model compression, quantization, and deployment on specialized accelerators. Integration & Interoperability o Work with O-DU stacks (FlexRAN, OAI, SRS, Aerial) to integrate & benchmark AI-RAN modules. o Ensure interoperability through open data interfaces that allow model insertion, data collection, performance measurement, and comparison across multiple parties. Research & Collaboration o Stay current on the latest ML and wireless research; assess applicability to 3GPP AI-RAN and ISAC use cases, contributing to research, development, and standardization efforts. o Collaborate with internal product teams and external partners (AI-RAN Alliance, 3GPP studies, OpenRAN Alliance) to drive adoption of AI-RAN modules and publish findings. o Contribute to publications, standardization, research items, conferences, and open-source software aligned with the OpenRAN and the Open AI-RAN vision. Minimum Qualifications: BS, MS, or PhD in Electrical/Computer Engineering, Computer Science, or related field Proficiency in at least one programming language (Python or C++ preferred) Familiarity with Deep Learning frameworks (e.g. – PyTorch, TensorFlow) Experience in some of the following areas: deep learning, RF Sensing, statistical signal processing/DSP, wireless/communications systems fundamentals, time/frequency analysis of signals, machine learning, channel estimation and equalization, MIMO systems, beamforming. Ability to work on open-ended and self-guided problems, building candidate solutions and coming up with appropriate metrics for comparison, system designs, and rigorous customer centric validation. Strong communication and teaming skills to work collaboratively and productively in a small company environment – allowing for broader and proactive scope of engagement across business functions. Qualifications of the ideal candidate: All the minimum qualifications, plus: Experience in building machine learning models optimized deployment hardware platforms Familiarity with communications systems and/or 3GPP 5G NR physical layer and ORAN architecture. Background in digital signal processing (multi-rate, polyphase filters, equalization, synchronization.) Experience in building & optimization ML / Deep Learning models for efficiency deployment Familiarity with widely used SW ML tools such as C++, CUDA, TensorRT, OpenCL, ONNX, etc. Experience in edge deployment of ML models and accelerator toolchains (TensorRT, CUDA, ONNX, Ryzen-AI, OpenCL, oneDNN, FPGA/AI implementation, etc). Prior work with Sionna PHY/RT, MATLAB, and other link and system level wireless tools/sims. Contributions to communications software or standards (e.g., AI-RAN Alliance, 3GPP, IEEE, etc.). Working at DeepSig DeepSig is growing its technical team while cultivating a collaborative, agile, and fun small-team culture. We value creativity, knowledge sharing, and employee growth, and we encourage participation in scientific publications, conferences, and open-source software. We offer competitive salaries and benefits, an employee stock option grant program, an environment where we are excited to be transforming and disrupting how signal processing is done with AI/ML, a welcoming and inclusive environment, a flexible schedule, and a great work / life balance. Equal Opportunity Statement: DeepSig is an equal opportunity employer and does not discriminate based on race, ethnicity, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability. We are dedicated to cultivating an inclusive, diverse, and engaging workplace where individuals feel fulfilled, inspired, and motivated. We value the unique perspectives that our team brings.

Responsibilities

The role involves designing and implementing AI/ML models for RAN functions and contributing to algorithm optimizations. Additionally, the engineer will develop dataset collection tools and integrate AI-RAN modules into existing systems.