Relevant Experience & Education Highlights
Expertise in building large-scale ML training infrastructures, GPU workload simulations for generative models like GANs, and PyTorch-based optimizations positions this candidate strongly for advancing Adobe's Firefly foundation model stack.
1. Build and optimize infrastructures that power large foundation model training on thousands of GPUs
Hands-on experience constructing scalable training systems and performance tools aligns directly with scaling generative AI workloads.
- Built infrastructure for large-scale training, experimentation, and performance visualizations at a major semiconductor company.
- Developed Python models simulating deconvolutional neural network architectures in a university research lab.
- Composed GPU simulations of GAN workloads, comparing against FPGA and CPU platforms to assess resource utilization.
- Deployed models improving compiler optimization techniques, heuristics, and autotuning processes using PyTorch.
2. Profile GPU utilization, trace inference and training runs and help craft strategies for optimizing our ML model latency
Proven ability to analyze GPU performance and quantify impacts on generative quality and latency supports optimization strategies.
- Quantified effects of design choices on generative quality, resource utilization, and computation latency in GAN GPU simulations.
- Leveraged data science pipelines generating actionable insights for partner teams at a major semiconductor company.
- Studied cutting-edge ML-based code optimization research to deploy performance-enhancing models.
3. Architect and optimize end-to-end ML pipelines, ensuring they're scalable, efficient, and robust
Background in data pipelines, Docker containerization, and cloud tools like AWS enables robust ML workflow development.
- Integrated AWS, Docker, HDFS, and PyTorch in building scalable ML experimentation pipelines.
- Applied exploratory data analysis, neural networks, and probabilistic graphical models across systems.
- Designed custom software solutions overcoming measurement challenges at a test and measurement company.
Requirements & Candidate Alignment
| Adobe Requirement | Candidate Qualification |
|---|---|
| Education: Graduate, PhD, or postgraduate degree in Computer Science, Computer Engineering, or a related field—or equivalent experience | MS and BS from top-tier research and public universities |
| Experience: 2+ years ML Engineering experience, specializing in generative AI like LLMs | 12.8 years total experience, including 3+ years ML engineering focused on generative models like GANs and large-scale training |
| Technical Skills: Strong Python and deep learning engineering skills, paired with experience in training and inferencing with PyTorch or TensorFlow | Advanced Python and PyTorch expertise: Built models for neural networks, VAEs, and large-scale training infrastructures |
| Domain Knowledge: Familiarity with distillation, transformers, and diffusion models. Experience with generative image and video is a plus | Generative AI proficiency: Simulated GAN workloads and deconvolutional networks, matching transformers and diffusion models |
| Deployment Technologies: Knowledge of deployment technologies such as Docker, ML Ops, and ML services | Docker and MLOps experience: Deployed scalable ML systems with Docker, Git, and Linux environments |
| Cloud Platforms: Experience with cloud platforms like Azure and AWS is a plus | AWS hands-on: Utilized AWS with Boto3, HDFS for data science and ML pipelines |
| Problem-Solving: Excellent problem-solving abilities and capacity to analyze complex issues and drive solutions with a data-driven approach | Data-driven optimization: Analyzed GPU simulations and pipelines to derive insights on performance and heuristics |
| Communication: Strong verbal and written communication skills and success in cross-functional team environments | Cross-functional collaboration: Advised on solution architectures and generated insights for partner teams |
If you would like to discuss this candidate or other critical roles, here is a link to my calendar to schedule a call:
Schedule a CallContact:
Finding the signal in the noise since 2005