About The Role In this role, you will combine deep hands-on expertise in GPU-accelerated AI, Generative AI, and modern AI infrastructure with the opportunity to shape enterprise AI strategies for integral clients. You will lead technical engagements from discovery through production, influence architectural decisions, drive AI adoption, and contribute to NVIDIA-focused go-to-market initiatives while collaborating with client stakeholders and multidisciplinary engineering teams.

Responsibilities Lead end-to-end AI engagements, from discovery and solution strategy through architecture design, implementation planning, and production delivery

Translate complex business challenges into AI use cases, solution roadmaps, and scalable enterprise architectures

Design and validate production-ready AI solutions leveraging NVIDIA technologies across cloud and on-premises environments

Define reference architectures for Generative AI and Agentic AI solutions, including GPU infrastructure, Kubernetes orchestration, and inference optimization strategies

Evaluate and optimize AI inference performance by analyzing GPU utilization, latency, throughput, scalability, and infrastructure efficiency

Benchmark and optimize LLM serving frameworks and deployment configurations

Lead pre-sales activities, including technical discovery, workshops, solution positioning, proposals, and proof-of-concept initiatives

Collaborate with NVIDIA stakeholders and internal teams to develop reusable accelerators, solution blueprints, and industry offerings

Drive technical thought leadership through whitepapers, technical content, conference presentations, and mentorship Requirements 6+ years of experience in AI consulting, Generative AI, Agentic AI, Machine Learning, or Deep Learning, including ownership of client-facing engagements

Bachelor's or Master's degree in Computer Science,



Applied Mathematics, Physics, Engineering, or related technical field preferred

Advanced expertise in Generative AI, Agentic AI, multimodal AI, transformers, Large Language Models (LLMs), and Vision Language Models (VLMs)

Hands-on experience with Python and modern AI/ML frameworks, including PyTorch, TensorFlow, Hugging Face, Pandas, and NumPy

Strong experience with NVIDIA AI technologies, including at least three of the following: NeMo, NIM, Triton, TensorRT-LLM, Riva, DeepStream, Metropolis, or Omniverse

Practical experience in deploying AI workloads on Kubernetes using Helm, NVIDIA GPU Operator, GPU device plugins, MIG/vGPU partitioning, and modern inference platforms such as vLLM or Ollama

Working knowledge of model quantization, inference optimization, and GPU profiling tools, including NVIDIA Nsight Systems, Nsight Compute, DCGM, PyTorch Profiler, and Triton or vLLM monitoring

Proven skill in analyzing GPU performance, identifying compute, memory, or I/O bottlenecks, and optimizing AI infrastructure for performance and cost efficiency

Experience in designing and deploying enterprise AI solutions on AWS, Azure, or GCP using CUDA, TensorRT, Triton Inference Server, DeepStream, and ONNX

Solid understanding of enterprise architecture, distributed systems, MLOps, AI governance, and modern software engineering practices

Strong advisory and stakeholder management skills

English proficiency for leading technical discussions with global clients and stakeholders SoftServe is an equal opportunity employer. Qualified applicants will receive consideration regardless of race, color, ancestry, ethnicity, national origin, religion, sex, sexual orientation, gender identity or expression, age, citizenship, disability, health condition, marital or family status, veteran status, or any other characteristic protected by applicable law.

📌 Lead Data Scientist (Nvidia) (Colombia)
🏢 SoftServe
📍 Colombia