AI/ML Engineer - Python

Company: NeuraNX.ai

Role: AI/ML Engineer - Python

Location: Chennai, Tamil Nadu

Experience: 3–5 Years

Work Mode: Hybrid

Employment Type: Full-time

Role Overview

As an AI/ML Engineer- Python, you will design, develop, deploy, and optimise AI/ML applications using Python, PyTorch, Hugging Face Transformers, and modern LLM inference frameworks. You will work on scalable inference APIs, model-serving pipelines, performance benchmarking, and production troubleshooting.

Key Responsibilities

• Develop AI and machine-learning applications using Python.
• Integrate, deploy, and serve open-source LLMs and multimodal models.
• Build scalable inference APIs using FastAPI, Flask, or similar frameworks.
• Work with Hugging Face Transformers, PyTorch, and model inference technologies.
• Implement streaming responses, asynchronous processing, and continuous batching.
• Optimise inference latency, throughput, GPU utilisation, and memory consumption.
• Work with model-serving frameworks such as vLLM, SGLang, NVIDIA Triton, TensorRT-LLM, KServe, or Ray Serve.
• Develop model benchmarking, evaluation, and load-testing pipelines.
• Monitor model performance, response latency, throughput, and inference reliability.
• Troubleshoot model-loading failures, GPU out-of-memory errors, and runtime bottlenecks.
• Write clean, modular, secure, maintainable, and well-tested Python code.
• Research and evaluate emerging AI models, inference frameworks, and optimisation techniques.

Required Skills and Qualifications

• 3–5 years of experience in Python development or AI/ML engineering.
• Strong programming skills in Python and experience building production-grade applications.
• Hands-on experience with PyTorch and Hugging Face Transformers.
• Practical experience deploying or serving large language models.
• Experience with FastAPI, Flask, or similar backend frameworks.
• Understanding of LLM fundamentals, including:
• Tokenisation, attention mechanisms, and context length
• KV cache and its impact on inference memory
• Model quantisation and memory optimisation
• Continuous batching and streaming inference
• Hands-on experience with at least one inference or model-serving framework: vLLM, SGLang, NVIDIA Triton, TensorRT-LLM, KServe, or Ray Serve.
• Experience developing REST APIs, WebSockets, or Server-Sent Events (SSE).
• Familiarity with PostgreSQL, Redis, or similar databases and caching technologies.
• Strong debugging, analytical, and problem-solving skills.
• Good communication, documentation, and collaboration skills.

Preferred Skills

• Experience working with NVIDIA GPUs and CUDA.
• Knowledge of FP16, INT8, INT4, AWQ, or GPTQ quantisation techniques.
• Experience with embeddings, reranking, OCR, speech AI, or vision-language models.
• Familiarity with model benchmarking, concurrency testing, and inference performance profiling.
• Experience developing Python SDKs, reusable AI libraries, or internal developer tools.
• Familiarity with CI/CD, containerisation, and production deployment practices.
• Contributions to open-source AI or machine-learning projects.

Education

• Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Machine Learning, Data Science, or a related discipline.
• Candidates with strong practical experience, demonstrable AI projects, and relevant open-source contributions will also be considered.

What We're Looking For

• A hands-on engineer who can turn AI models into reliable, production-ready applications.
• Strong ownership across design, implementation, debugging, and optimization.
• Ability to independently research, evaluate, and integrate open-source AI technologies.
• A performance-oriented mindset with an interest in scalable inference systems.
• A willingness to experiment, learn, and solve complex engineering challenges.

Note: Application Process

Interested candidates are requested to share their updated resume along with:

• GitHub profile or portfolio link
• Details of relevant AI/ML or Python deployment projects
• Current location
• Notice period
• Current CTC and expected CTC

Please apply through LinkedIn or share your profile with our hiring team – [HIDDEN TEXT]

Join NeuraNX.ai and help build efficient, scalable, and production-ready AI systems.