Lead ML Engineer - LLM Inference & NLP
Social Discovery Group1 month ago
Remote, SerbiaStaff+
Responsibilities
- Optimize and scale production LLM inference using caching, batching, quantization, speculative decoding, and related techniques.
- Run distributed inference for very large models across multi-GPU and multi-node environments.
- Benchmark new GPU servers and hardware, integrate them into production, and adapt serving code.
- Provide technical leadership to NLP and computer vision teams through reviews, mentoring, direction-setting, and technical decisions.
- Train and fine-tune language models, including post-training, and improve agent harnesses and chat algorithms.
- Track inference and post-training research and open-source developments and translate them into the ML roadmap.
- Collaborate with validation, content, and dataset preparation teams to design experiments and measure model quality.
Requirements
- Deep hands-on production experience optimizing LLM inference with SGLang, vLLM, or TensorRT-LLM.
- Experience with distributed inference or training of large models, including MoE, tensor, expert, or pipeline parallelism and multi-node GPU clusters.
- Strong understanding of KV caching, attention kernels, batching, quantization, and GPU profiling.
- Experience training and fine-tuning LLMs, including RLHF, DPO, or similar post-training methods.
- Proven technical leadership through engineering reviews, mentoring, and technical decisions while continuing to write code.
- Proficiency with PyTorch, transformers, and related libraries.
- Experience at an AI-focused startup or company is a strong plus.
- Backend engineering experience with Python, Go, or C# and knowledge of scalable deployment systems is a significant advantage.
- Advanced English or Russian.
- CUDA or Triton kernel development, computer vision, multimodal LLMs, first-author papers, and notable open-source contributions are nice to have.
- A degree in computer science, mathematics, or physics from a strong program, such as an MSc or PhD, is nice to have.
Benefits
- Fully remote, full-time opportunity.
- 28 calendar days of vacation and 7 wellness days per year.
- Up to $5000 in bonuses for recommending successful applicants.
- 50% payment for professional training, international conferences, and meetings.
- Corporate discount for English lessons.
- Health benefits, including up to $1,000 gross per year for eligible employees who cannot use corporate medical insurance.
- Equipped workplace and necessary equipment in offices or coworking locations, or up to $1,000 gross reimbursement every three years for remote workplace costs.
- Internal gamified gratitude system with merchandise, team-building activities, and massage certificates.
Categories
About Social Discovery Group
Social Discovery Group builds and operates consumer social-discovery and entertainment platforms that help people meet and interact online, and it also invests in related IT startups. The company monetizes its products as consumer internet services and runs multiple brands in dating, live-streaming, and chat experiences. Founded in 1991 and headquartered in Malta, it is privately held and works with distributed teams serving users in multiple countries.