Senior Software Engineer - ML Systems / AI Frameworks / Compilers

Key Skills

Aartificial intelligence, C++, Deep Learning, GPUs, Large Language Models - LLMs, Machine learning techniques, Optimization, Python Programming, SDK, Software Development

Job Description

We’re looking for a skilled software engineer to join our AI SDK team, building a highly flexible Python software development kit to quantize and optimize models for inference on Recogni’s hardware accelerators. You will help architecting this library from ground-up, focusing on the intersection between the compiler and the Python SDK / ML framework. If that matches your experience and interest, we would love to talk!

Where you’d help us:

• AI Stack Development: Architect and shape our SDK software stack, allowing developers to deploy state-of-the-art generative AI models (LLMs, stable diffusion, …) to our hardware. Your focus area will be the deep learning compiler frontend and the graph intermediate representation (IR).
• Compiler Synergies: Closely collaborate with our compiler team to design an API that enables developers to manually optimize algorithms on our hardware.
• Performance Optimization: Deeply analyze state-of-the-art AI networks and optimize them for Recogni’s accelerator by implementing hardware-specific kernels.
• Distributed Inference: Build the SW infrastructure around sharding and collectives that lets developers seamlessly deploy large deep learning models for distributed inference.
• HW/SW Co-design: Collaborate closely with AI scientists to investigate the most recent advances in machine learning, analyze their runtime on our hardware, and contribute to the hardware-software co-design of our next generation.

Qualifications:

• Experience: 4+ years of relevant software engineering experience.
• Programming Languages: Proven proficiency in Python and ideally C++.
• ML Frameworks: Experience in AI training/inference framework development OR familiarity with the export mechanisms, operator sets, and intermediate representations of popular ML frameworks such as PyTorch.
• [Preferred] ML System Optimization: Experience in optimizing ML systems based on runtime analyses of latency, memory bandwidth, I/O access, and compute utilization.
• [Preferred] Distributed Systems: A good understanding of and ideally experience with high-performance distributed computing.
• [Preferred] Deep Learning Compilers: Experience with MLIR, LLVM, IREE, XLA, TVM, or Halide is a plus.
• [Preferred] Hardware Affinity: Knowledge of GPU, CPU, or AI hardware accelerator architectures.

Reasons to consider joining Recogni:

• Ground floor opportunity with the team; be part of shaping one of the most exciting new AI products.
• Learning and development opportunities from a highly diverse and talented peer group, including experts in a wide range of fields, from Artificial Intelligence to Systems & Device Engineering.
• Perks including meals, snacks, drinks and us!
• Sharp, motivated co-workers in a fun office environment
• Employee Stock Purchase Plan
• Flexible work hours & generous PTO policies

Recogni is an equal opportunity employer. We believe that a diverse team is better at tackling complex problems and coming up with innovative solutions. All qualified applicants will receive consideration for employment without regard to age, color, gender identity or expression, marital status, national origin, disability, protected veteran status, race, religion, pregnancy, sexual orientation, or any other characteristic protected by applicable laws, regulations and ordinances.

A note to Recruitment Agencies: Please don’t reach out to Recogni employees or leaders about our roles -- we’ve got it covered. We don’t accept unsolicited agency resumes and we are not responsible for any fees related to unsolicited resumes. Thank you for your understanding.