AI Engineer

We are looking for an experienced AI Engineer who can independently understand business problems, select the right technical approach and build reliable AI solutions from experimentation through production.

This is a hands-on role requiring strong engineering fundamentals, systematic problem-solving and ownership of solution quality, performance and scalability. You will work closely with Product, Delivery, Engineering, QA and client-facing teams.

Key Responsibilities

• Own end-to-end AI development: Take solutions through research, Proof of Concept, development, testing, UAT, production deployment and ongoing improvement.
• Build ML and Deep Learning solutions: Develop pipelines covering data preparation, feature engineering, model selection, training, optimization, validation, monitoring and retraining.
• Develop Generative AI applications: Build RAG, knowledge retrieval and agentic solutions using prompt engineering, embeddings, vector databases, structured outputs and tool calling.
• Select and optimize models: Evaluate open-source and commercial LLMs, SLMs and classical ML models based on accuracy, latency, privacy, cost and business requirements. Apply fine-tuning and optimization techniques where appropriate.
• Validate AI quality: Create benchmark datasets, automated evaluations and regression tests to assess accuracy, relevance, groundedness, completeness, consistency and hallucination.
• Engineer for production: Address concurrency, batching, caching, asynchronous processing, GPU utilization, throughput, failure handling and infrastructure cost.
• Implement MLOps and LLMOps: Establish versioning, experiment tracking, CI/CD, controlled deployments, observability, drift detection and rollback practices.
• Build application services: Develop robust Python backend services, APIs and integrations, with sufficient frontend capability to independently demonstrate end-to-end AI prototypes.
• Investigate and resolve failures: Debug issues across data, models, prompts, retrieval, APIs, infrastructure and application logic, identifying and addressing root causes.
• Collaborate and contribute: Participate in architecture discussions, communicate risks and dependencies, support engineers and research emerging AI approaches.

Required Skills

• Strong Python proficiency and experience writing clean, maintainable, testable production code.
• Strong foundations in Machine Learning, Deep Learning, NLP and Transformers, with hands-on experience in PyTorch or TensorFlow and Hugging Face.
• Practical experience with LLMs, SLMs, RAG, embeddings, vector databases, prompt engineering and LLM evaluation.
• Understanding of model architectures, tokenization, attention, context management, inference behaviour and model limitations.
• Experience with model adaptation and optimization approaches such as LoRA, QLoRA, PEFT, instruction tuning, quantization or distillation.
• Backend development using FastAPI, Flask or Django, with REST APIs, databases, asynchronous processing, authentication, RBAC and third-party integrations.
• Working knowledge of SQL/PostgreSQL, Git, Docker and Linux.
• Ability to independently review, debug and take ownership of implementations, including code generated using AI-assisted tools.

Preferred Skills

• AI orchestration: LangGraph, LangChain or equivalent frameworks.
• Model serving and infrastructure: vLLM, Triton, TGI, TensorRT-LLM, GPU deployment, Kubernetes and cloud, private cloud or on-premise environments.
• MLOps and evaluation: MLflow, Kubeflow, Airflow, DVC, RAGAS, DeepEval, LangSmith, promptfoo or equivalent tools.
• Application development: Redis, Celery, WebSockets, React/Next.js, JavaScript/TypeScript and HTML/CSS.
• Exposure to Speech-to-Text, Conversational AI, Voice AI, Computer Vision, Recommendation Systems or Multimodal AI.

Experience & Working Approach

• Demonstrated ownership of AI solutions from PoC to production; experience supporting real production users is strongly preferred.
• Strong research, debugging and communication skills, with the ability to work independently and collaborate across teams.
• Candidates with different experience levels may be considered where they demonstrate exceptional technical depth and ownership.