12 Oct
|
Virtusa
|
Hyderabad
LLM System Architecture, Supervised Fine-Tuning (SFT), Knowledge Distillation, Long-Context Mitigation (>15k Tokens), MLOps Architecture Design, vLLM Inference Optimization, INT8 Quantization Strategy
Model Evaluation Harness Architecture, Benchmark Analysis (Qwen3/Llama 3.1/Mistral-Nemo), Token Cost Trade-off Modeling, Responsible AI Framework Evaluation, Shadow Mode Divergence Strategy
📌 Architect (Hyderabad)
🏢 Virtusa
📍 Hyderabad