Lead the design and development of scalable, real-time AI systems at a Chennai-based technology solutions company. This role involves architecting end-to-end solutions from model development to production deployment, with a focus on low-latency inference and MLOps best practices.

About the Company

Recode Solutions is a technology-driven organization based in Chennai, Tamil Nadu, India. The company specializes in building advanced software and AI-powered systems for diverse business domains. With a focus on innovation and scalability, Recode Solutions partners with clients to deliver production-ready solutions that leverage the latest in artificial intelligence, machine learning, and real-time data processing. The company fosters a collaborative culture where engineers, data scientists, and product teams work together to solve complex problems.

Role Overview

We are seeking an experienced AI/ML Architect to lead the design and development of scalable, real-time AI systems. This is a full-time, senior-level position (10-12 years of experience) based in Chennai. The successful candidate will work closely with product, data, and engineering teams to architect end-to-end solutions — from model development and deployment to system integration and production monitoring. The role demands deep technical expertise in building low-latency inference pipelines, selecting appropriate infrastructure, and ensuring reliability and observability of deployed systems. You will also mentor engineers and help establish best practices for MLOps across the organization.

Key Responsibilities

• Design and architect AI/ML systems that are scalable, low-latency, and production-ready.
• Lead the development of real-time inference pipelines for use cases like voice, vision, or NLP.
• Select and integrate appropriate tools, frameworks, and infrastructure (e.g., Kubernetes, Kafka, TensorFlow, PyTorch, ONNX, Triton, VLLM).
• Collaborate with data scientists and ML engineers to productionize models.
• Ensure reliability, observability, and performance of deployed systems.
• Conduct architecture reviews, proof-of-concepts (POCs), and system optimizations.
• Mentor engineers and help set best practices for the ML lifecycle (MLOps).
• Drive the adoption of containerized environments and cloud-native architectures for ML workflows.
• Stay current with emerging technologies in AI/ML and recommend improvements.

Tech Stack

• Languages: Python, Bash
• ML/DL Frameworks: TensorFlow, PyTorch, scikit-learn
• LLM Inference Optimization: vLLM, TensorRT-LLM, SGLang
• Infrastructure & Deployment: Kubernetes, Docker, Kafka, ONNX, Triton Inference Server
• Data & Retrieval: Vector databases, embedding-based retrieval, RAG pipelines
• Cloud Platforms: AWS, GCP, or Azure
• CI/CD & MLOps: CI/CD practices for ML workflows
• Operating System: Linux (Ubuntu)
• Other: Event-driven pipelines, microservices, distributed training/inference, model compression/optimization, edge deployments

6+ years of experience building and deploying ML systems in production. Proven expertise in real-time, low-latency system design (streaming inference, event-driven pipelines). Strong understanding of scalable architectures — microservices, message queues, distributed training/inference. Proficient in Python and ML/DL frameworks (scikit-learn, TensorFlow, PyTorch). Hands-on experience with LLM inference optimization using vLLM, TensorRT-LLM, and SGLang. Familiarity with vector databases, embedding-based retrieval, and RAG pipelines. Experience with containerized environments (Docker, Kubernetes) and managing multi-container applications. Working knowledge of cloud platforms (AWS, GCP, or Azure) and CI/CD practices for ML workflows.

Opportunity to work on cutting-edge AI systems and real-time inference pipelines. Collaborate cross-functionally with product, data, and engineering teams in a dynamic environment. Mentorship opportunities and the chance to set best practices for ML lifecycle (MLOps). Exposure to edge deployments and model compression/optimization techniques.