AI Architect

Ethics Infotech; Vadodara; On-site; INR 2,500,000 - 4,000,000; Full time

Job summary

Ethics Infotech in Vadodara invites a hands-on Technical Engineering Manager to lead AI/ML and Computer Vision initiatives. You will architect CV pipelines, guide cross-functional teams, and deliver production-grade solutions across edge, on-prem, and cloud environments.

The role requires strong expertise in CV, ML, and GenAI, with leadership skills to mentor engineers, make cross-team decisions, and drive reliable business outcomes.

Qualifications

• 6+ years of software and AI/ML engineering experience.
• 2+ years in technical leadership or engineering management.
• Deep CV expertise with YOLOv5/v8, R-CNN, DeepSORT and edge deployment.

Responsibilities

• Architect end-to-end CV pipelines and GenAI/LLM-powered systems tailored to multi-domain problems.
• Define technical standards for data ingestion, annotation, augmentation, model evaluation, MLOps, and production serving.
• Lead cross-team architectural discussions spanning software engineering, embedded hardware/edge devices, and cloud infrastructure.

Skills

Computer Vision

Machine Learning

GenAI

Edge Deployment

Python

PyTorch

TensorRT

OpenCV

C++

LLMs

Education

Bachelor’s or Master’s in CS/AI/EE

Tools

OpenCV

NumPy

TensorRT

OpenVINO

ONNX

Docker

PyTorch

Job description

Roles And Responsibilities

Architectural Leadership & System Design

Skills

Job Location: Vadodara Office Hours: 09:30 am to 7 pm Experience: 6+ Years We are seeking a hands-on, forward-thinking Technical Engineering Manager – AI, Machine Learning & Computer Vision to drive our applied AI and intelligent vision initiatives. In this role, you will lead end-to-end technical execution: architecting complex CV and ML systems, guiding cross-functional teams from concept to production, and building production-grade solutions across edge, on-prem, and cloud environments. The ideal candidate combines deep engineering expertise in Computer Vision (object detection, video analytics, OCR), foundational Machine Learning, and Generative AI (LLMs, multimodal architectures) with the leadership ability to mentor engineers, make cross-team technical decisions, and deliver reliable solutions to business problems. Role & Responsibilities

Roles And Responsibilities

• Architect and oversee end-to-end Computer Vision pipelines, statistical ML algorithms, and GenAI/LLM-powered systems tailored to multi-domain business problems.
• Define technical standards for data ingestion, annotation, augmentation, model evaluation, MLOps, and production serving.
• Lead cross-team architectural discussions and make key technical decisions that span software engineering, embedded hardware/edge devices, and cloud infrastructure.

Hands-on Technical Execution & Delivery

• Actively develop, train, evaluate, and benchmark deep learning models for object detection, multi-object tracking, image classification, semantic segmentation, and anomaly detection.
• Build and optimize real-time image/video processing pipelines and edge deployments using TensorRT, Open VINO, or ONNX.
• Drive the implementation and integration of Large Language Models (LLMs) and multimodal AI techniques into existing workflows, applications, and dashboards.
• Guide the team through robust code reviews, continuous testing, data drift mitigation, and model performance tuning.

Team Leadership & Project Management

• Guide and mentor a multidisciplinary team of CV engineers, ML developers, and data scientists across sprint planning, technical roadblocks, and career growth.
• Champion Agile delivery cadences (e.g., bi-weekly releases, milestone reviews) in close coordination with the PMO and engineering leads.
• Collaborate closely with Talent Acquisition to interview, assess, and onboard top-tier AI and computer vision engineering talent.

Education/Qualification (if Any Certification)

• Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Electrical Engineering, Data Science, or a related quantitative field.

Requirements

• 6+ years of practical software and AI/ML engineering experience, including 2+ years in a technical leadership, lead, or engineering management role.
• Deep practical experience with OpenCV, NumPy, and modern CV architectures (YOLOv5/v8, Mask R-CNN, Faster R-CNN, DeepSORT).
• Demonstrated track record with at least one real-world, real-time CV domain (e.g., surveillance, physical security/access control, retail analytics, or industrial defect inspection).
• Strong foundation in image processing concepts (thresholding, contour analysis, spatial transformations, feature extraction).
• Solid grounding in statistical analysis and core ML algorithms (regression, tree-based models, clustering).
• Hands-on experience with deep learning frameworks: PyTorch or TensorFlow.
• Practical proficiency in Large Language Models (LLMs), prompt engineering, and multimodal AI integrations.
• Proven experience optimizing models via quantization and pruning for edge deployment (e.g., NVIDIA Jetson Nano/Xavier/Orin, Coral, or Intel edge hardware).
• Experience with runtime acceleration tools like TensorRT, OpenVINO, or ONNX Runtime.
• Advanced proficiency in Python (knowledge of modern C++ is a strong plus).

Solid understanding of REST/gRPC APIs, microservices, containerization (Docker), and automated CI/CD pipelines.

Good To Have

• Document Intelligence & OCR: End-to-end document processing pipelines (PaddleOCR, Tesseract, AWS Textract, Google Vision API), table/form extraction, and Named Entity Recognition (NER).
• High-Throughput Video Analytics: Familiarity with NVIDIA DeepStream, GStreamer, or interactive front-end tools (Streamlit, Gradio).
• MLOps & Infrastructure: Experience with Triton Inference Server, MLflow, Kubeflow, and cloud AI platforms (AWS SageMaker, Azure ML, or GCP Vertex AI).
• Hardware Interfacing: Working knowledge of camera sensors, RTSP streaming, CUDA programming, and GPU acceleration.
• Research & Publications: Background in academic CV research or published papers in recognized conferences/journals (CVPR, ICCV, ECCV, NeurIPS).