Senior AIML Engineer (LLM, RAG, Computer Vision & MLOps)

Role Overview

We are seeking a highly skilled Senior AI/ML Engineer to design, develop, and deploy production-grade AI systems that combine Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Computer Vision, and scalable MLOps infrastructure.

In this role, you will build end-to-end AI pipelines that transform unstructured documents and video streams into structured insights, validated outputs, and intelligent analyses. You will work across the full lifecycle—from data ingestion and model development to deployment, monitoring, evaluation, and optimization in production environments.

The ideal candidate combines strong machine learning and software engineering fundamentals with hands-on experience building reliable AI systems that balance accuracy, latency, scalability, cost, and determinism.

Key ResponsibilitiesAI & LLM Systems

• Design and implement production-grade RAG pipelines including document ingestion, preprocessing, chunking, embedding generation, vector retrieval, and LLM-powered structured extraction.
• Develop structured JSON extraction systems using schema-based approaches (e.g., Pydantic), function calling, JSON mode, and deterministic validation layers.
• Build and optimize retrieval strategies including chunking methodologies, metadata filtering, reranking, multi-query retrieval, deduplication, and retrieval evaluation.
• Integrate state-of-the-art LLMs to generate contextual analyses, explanations, recommendations, and domain-specific insights from structured and unstructured data.
• Design hybrid AI architectures that combine deterministic rules, heuristics, and domain knowledge with probabilistic AI models.
• Implement robust anti-hallucination mechanisms including context-grounded prompting, validation workflows, confidence scoring, and evidence traceability.

Computer Vision & Video Analytics

• Design and optimize computer vision pipelines for processing images and video streams.
• Develop real-time or batch-based human pose estimation, tracking, and landmark extraction systems.
• Build feature engineering pipelines that convert visual signals into meaningful biomechanical, spatial, or behavioral metrics.
• Develop algorithms for calculating motion characteristics, geometric relationships, temporal patterns, and other derived analytical features.
• Optimize video processing workflows using OpenCV, FFmpeg, frame sampling, and inference acceleration techniques.

Data Processing & Validation

• Build scalable ingestion pipelines for PDFs, DOCX, XLSX, images, and video data.
• Implement OCR-based extraction workflows for scanned or image-based documents.
• Develop deterministic post-processing and normalization layers for extracted data.
• Create validation frameworks that enforce schema compliance, business rules, cross-field consistency, and output correctness.
• Design evaluation pipelines and benchmark datasets to continuously measure extraction quality and model performance.

MLOps & Production Engineering

• Build, deploy, and maintain production AI services using API-first architectures.
• Develop scalable microservices using FastAPI, Flask, or similar frameworks.
• Containerize and deploy AI workloads using Docker and cloud-native infrastructure.
• Manage model deployment, monitoring, observability, logging, and performance optimization.
• Implement CI/CD workflows, model versioning, testing, and production release processes.
• Ensure secure handling of customer data, tenant isolation, and compliance with enterprise-grade deployment requirements.

Required QualificationsMachine Learning & AI

• Strong understanding of machine learning fundamentals and production AI system design.
• Hands-on experience with modern LLMs (OpenAI, Anthropic, or equivalent) and prompt engineering for structured and contextual outputs.
• Practical experience building RAG-based applications using embeddings, vector search, and retrieval optimization techniques.
• Experience with vector databases such as ChromaDB, Pinecone, Weaviate, FAISS, or similar platforms.
• Experience designing hybrid systems that combine deterministic logic with AI-driven reasoning.

Computer Vision

• Experience building computer vision pipelines using frameworks such as MediaPipe, OpenPose, YOLO-Pose, or equivalent.
• Proficiency with PyTorch and/or TensorFlow.
• Strong understanding of video processing, tracking, image transformations, and inference optimization.
• Hands-on experience with OpenCV and related video processing libraries.

Software Engineering

• Advanced Python programming skills with a strong focus on clean, maintainable, production-quality code.
• Experience building backend services using FastAPI, Flask, or similar frameworks.
• Strong understanding of API design, data validation, serialization, and service architecture.
• Experience working with JSON schemas, Pydantic, and defensive programming patterns.

Cloud & MLOps

• Experience deploying AI/ML systems on AWS, GCP, Azure, or equivalent cloud platforms.
• Strong Docker and containerization expertise.
• Familiarity with scalable inference architectures, monitoring, logging, and performance tuning.
• Experience building and maintaining production ML pipelines and deployment workflows.

Data & Analytics

• Strong proficiency with NumPy, Pandas, SciPy, and related scientific computing tools.
• Experience developing evaluation frameworks for AI systems, including accuracy measurement, regression testing, and output validation.
• Ability to balance model accuracy, latency, scalability, reliability, and operational cost.

Preferred Qualifications

• Experience with biomechanical analysis, sports analytics, motion tracking, or action recognition systems.
• 3+ years of hands-on experience in AI/ML engineering roles (or equivalent demonstrated impact).
• Experience implementing multi-stage extraction pipelines, refinement workflows, and validation passes.
• Familiarity with OCR systems and document intelligence platforms.
• Experience building visual comparison, side-by-side analysis, or explainable AI interfaces.
• Knowledge of inference optimization frameworks such as ONNX Runtime, TensorRT, or equivalent acceleration technologies.
• Experience with multi-tenant AI platforms and enterprise-scale deployments.
• Familiarity with AI evaluation methodologies, golden datasets, precision/recall analysis, and prompt regression testing.