Principal AI Engineer

You will be a hands-on technical leader responsible for designing, building, and deploying production-grade AI systems across data intelligence, computer vision, and generative AI domains. You will operate at the intersection of applied AI and industrial-scale deployment, owning end-to-end AI solutions that directly power autonomous operations, predictive intelligence, and workforce augmentation. This role combines deep technical ownership, system-level thinking, applied frontier AI, and close collaboration with data platform, IoT, and business teams. The core responsibilities for the job include the following: Generative AI and Agentic Systems for the Enterprise: • Build enterprise-grade GenAI systems that are safe, reliable, and deeply embedded in operations. • Design and implement firm-wide multimodal GenAI capabilities over enterprise data. • Build RAG-powered assistants, copilots, and AI agents for operations, engineering, and field teams. Develop AI agent meshes that: • Assist field engineers in day-to-day operations. • Orchestrate workflows across enterprise systems. • Reason over telemetry, images, documents, SOPs, and knowledge bases. • Ensure robustness, observability, governance, and responsible-AI practices in production GenAI systems. • Human-in-the-loop AI systems for safety-critical and compliance-heavy workflows 2 (Secondary). Data Intelligence at Massive Scale: • Build AI systems over one of the largest infrastructure telemetry footprints in the country. • Ingest and model terabytes of telemetry data daily from ~250,000+ sites across India. Design ML systems to: • Detect hidden patterns and anomalies in real time. • Generate intelligent alerts with minimal noise. • Predict outages and failures before they occur. • Forecast energy and resource consumption. • Predict asset degradation and lifecycle risks. • Automatically determine root causes (RCA) across complex systems 2 (secondary). Industrial Computer Vision (Applied CV): • Apply cutting-edge computer vision/VLMs to real-world infrastructure and operations challenges. • Detect defects and risks in tower and infrastructure components. • Architect and optimize low-latency, low-footprint CV models for mobile phones. • Enable live AI video for visual inspection, training, and skill augmentation. • Deploy and scale drone-based inspection systems for remote and hard-to-reach sites. Your Impact Move Indus: • Towers from reactive operations predictive autonomous. • Reduce downtime, cost, and manual intervention at a national scale. • Empower field teams with AI copilots and intelligent assistance. • Build AI systems that are used every day in production, not proofs of concept. Requirements: • 8-12 years of experience building and deploying AI/ML systems. • Strong hands-on experience in generative AI. • Proven ability to take AI systems from prototype to production at scale. • Experience working with large-scale data and complex, operation-heavy systems • Experience operating in 01 or high-ambiguity environments. Technical Skills: • Strong programming skills in Python (and related ML/data ecosystems) AI agent creation and orchestration: Designing tool-using agents, multi-agent workflows, routing, and planning vs. reactive patterns. • Experience with Vision language models. • Building and maintaining agent skills/tools (APIs, function calling, agent workflows, skills, and agent memory) aligned to operational use cases. • Model adaptation (hands-on): Fine-tuning and distillation for domain tasks. • Evaluation harnesses for quality, safety, latency, and cost of MCP servers/tool backends: Setting up and operating MCP servers to expose enterprise tools/services to agents. Defining schemas, permissions, and observability for tool execution. • Optional grounding in time-series modeling, anomaly detection, and forecasting. Computer vision pipelines (training, evaluation, deployment). Leadership and Mindset: • Builder mindset with a strong execution bias. • Operates as a technical owner, not just a coordinator or advisor. • Comfortable navigating ambiguity and shaping problem statements. • Able to communicate complex AI concepts to non-technical stakeholders. • Passionate about applying frontier AI to real operational problems.