AI Systems EngineerKodense
Kodense builds private cloud AI infrastructure and forward-deployed engineering teams for enterprise environments. You’ll be working directly on production AI systems—no wrapper apps, no slow corporate bureaucracy.
What You'll Work On:
• High-Throughput Inference: Deploy and optimize open-weight models (Llama 3, Mistral, DeepSeek) in production using vLLM, Ollama, and Triton.
• Private Cloud Infrastructure: Set up secure, isolated VPC environments and GPU compute instances with zero public data leakage.
• Low-Latency AI APIs: Build production Python & FastAPI microservices with real-time token streaming, rate limiting, and 99.9% uptime.
• Model Optimization: Fine-tune latency and memory footprint using quantization (AWQ/GGUF), KV caching, and GPU memory tuning.
What We Look For:
• 3+ years of production experience in Python, Docker, Linux, and backend API engineering.
• Hands-on experience serving LLMs in production (vLLM, Hugging Face, or Triton).
• Familiarity with cloud environments (AWS, GCP, or Azure) and private networking.
• Strong systems fundamentals (concurrency, memory management, and debugging).
Compensation & Benefits:
• ₹7,00,000 Fixed CTC (100% fixed cash: ₹58,333/month, zero withheld variable).
• ~98% take-home salary under Section 115BAC near-zero tax threshold.
• Up to ₹30,000 notice period buyout support.
• Hybrid work model at Prestige Cube, Koramangala, Bengaluru.
• Fast 48-hour hiring turnaround with direct founder review.
Apply directly in 2 minutes:
https://kodense.ai/careers
tal saysbuild private cloud ai infrastructure and optimize high-throughput production llm inference systems in bengaluru
ai-ml
mid
hybrid
Bangalore
pythonvllmfastapidockertriton
comments 1 anon