Job Title

Job Title: Machine Learning EngineerCompany: Dataism ServicesLocation: Viman Nagar, Pune, Maharashtra, IndiaExperience: 4+ YearsEmployment Type: Full-TimeRole OverviewWe are hiring a Machine Learning Engineer to own the path from trained model to production system. You will build inference pipelines, optimize model serving, integrate ML outputs into inspection planning dashboards, and ensure AI recommendations are reliable, explainable, and performant. You will also contribute to training infrastructure for fine-tuning and pretraining workflows on domain-specific data.Key ResponsibilitiesBuild and maintain production ML inference pipelines with REST/gRPC APIs, model versioning, and automated validation gates.Optimize models for production deployment using quantization, distillation, mixed-precision inference, and efficient batching strategies.Design and implement data pipelines that feed training and inference workflows, handling industrial-scale data with NULL handling, schema inconsistencies, and cross-system joins.Deploy and manage cloud infrastructure for GPU workloads including instance selection (T4, A10G, A100), cost optimization, and capacity planning on AWS.Integrate LLM-powered features into production applications: RAG pipelines, embedding services, prompt orchestration, and token-cost management.Implement MLOps best practices: model registries, A/B testing, monitoring for drift and degradation, and automated retraining triggers.Build React + Node.js dashboards and tools that surface AI outputs to end users in inspection planning workflows.Collaborate with data scientists to productionize research models with minimal performance regression.Essential Skills4+ years of professional experience in ML Engineering, Backend Engineering, or MLOps.Strong software engineering fundamentals in Python with clean, testable, well-documented code. Experience with a systems language (Rust or C++) is a strong plus.Hands-on experience deploying ML models to production: Docker, CI/CD pipelines, REST API design, and model serving frameworks (TorchServe, Triton, or similar).Deep familiarity with the PyTorch ecosystem: model optimization, fine-tuning transformer architectures, debugging training issues, and mixed-precision workflows.Strong cloud infrastructure experience (AWS preferred): EC2 GPU instance selection, SageMaker endpoints, S3 data pipelines, Bedrock for managed LLM deployment, and cost-aware architecture.Proficiency with SQL Server and relational databases in production: data pipeline construction, NULL handling, VARCHAR(MAX) patterns, async fetch, and connection pooling.Experience with MLOps tooling: model registries, experiment tracking, monitoring for data drift and model degradation, and automated retraining.Working knowledge of LLM integration patterns: prompt engineering, RAG, embedding pipelines, fine-tuning workflows, and token-aware cost management.Familiarity with frontend technologies (React, Node.js) sufficient to build AI-powered dashboards and internal tools.Ability to work fast, independently, and efficiently in a startup-paced environment.Preferred QualificationsExperience with Rust for backend services, ODBC integrations, or systems-level ML tooling.Familiarity with industrial inspection domains: PCMS, RBI (API 580/581), NDT methods, or petroleum refinery workflows.Experience with sensor data pipelines and IoT integration for condition monitoring or equipment health assessment.Knowledge of Azure infrastructure: VM provisioning, Event Grid monitoring, and GPU inferencing comparisons.