ML Engineer I_Python, Generative AI, FastAPI, ML

Role description

ML Engineer I AI / ML Engineer | Generative AI, Agentic AI & Enterprise ML

Who We Are

Born digital, UST transforms lives through the power of technology. Our 29,000+ practical problem solvers and creative thinkers work across more than 30 countries, helping clients build human-centered solutions and lasting business impact. Our teams combine innovation, agility, and engineering excellence across healthcare, financial services, insurance, retail, manufacturing, telecommunications, and other industries.

You Are

You are an AI-first engineer who enjoys turning ambiguous customer challenges into working, production-grade AI systems. You bring strong Machine Learning fundamentals, hands-on Generative AI experience, and the ability to build reliable agents, RAG applications, and cloud-native services. You move quickly, learn new tools deliberately, and take ownership from proof of concept through measurable customer outcomes.

The Opportunity

Builder Mode

• 0→1 engineering. Turn ambiguous requirements into working AI solutions: define the scope, select the right models and tools, build a usable first version, and iterate with customer feedback.
• Production ownership. Design, develop, deploy, and optimize ML and Generative AI applications using Python, modern ML frameworks, and secure cloud services.
• Agentic systems. Create intelligent agents and multi-agent workflows with tool calling, planning, state management, memory, context engineering, and reliable orchestration.
• RAG from data to answer. Own document ingestion, parsing, chunking, embeddings, vector indexing, hybrid retrieval, reranking, and grounded response generation for enterprise RAG.
• Backend integration. Develop FastAPI or Flask services, REST APIs, and integrations with enterprise data sources, microservices, and event-driven systems.
• AI-first delivery. Use Claude Code, Codex, Cursor, or similar tools to accelerate development, automate repetitive work, and improve the quality of tested, maintainable code.

Firefighter Mode

• A customer needs a working AI capability under a tight deadline. You identify the smallest viable solution, build it, and make the tradeoffs explicit.
• An agent is failing, a RAG answer is unsupported, or inference latency spikes. You reproduce the issue, inspect traces and metrics, isolate the cause, and close the loop with a measurable fix.
• A new model or framework promises better performance. You run a focused evaluation against quality, latency, cost, and security requirements before recommending adoption.
• You move into unfamiliar customer environments, collaborate with engineers and stakeholders, and communicate risks, blockers, and outcomes without waiting for perfect requirements.

Systems Thinker Mode

• Design reusable agent and RAG architectures that separate orchestration, tools, retrieval, model serving, application services, and enterprise integrations.
• Build evaluation and observability into the system: groundedness, retrieval quality, task success, accuracy, latency, token usage, cost, and failure analysis.
• Develop ML pipelines spanning data preparation, feature engineering, training, validation, inference, monitoring, and continuous improvement.
• Apply fine-tuning and optimization techniques such as PEFT, LoRA, instruction tuning, and domain adaptation when evidence supports them.
• Engineer for scalability, resilience, secure access, data protection, governance, and Responsible AI; turn successful prototypes into maintainable cloud-native services.

Minimum Qualifications

• Experience. 3–5 years of hands-on experience in ML, AI, or Generative AI engineering, including substantial project ownership and experience delivering enterprise applications into production.
• Education. A Master’s degree in Computer Science, AI, ML, Data Science, Engineering, or a related field is preferred. A Bachelor’s degree with exceptional relevant experience will also be considered.
• Core engineering. Strong Python skills and practical knowledge of supervised and unsupervised learning, neural networks, deep learning, transformers, NLP, statistical modeling, and model evaluation.
• Applied AI. Hands-on experience with at least one agent framework, production-oriented RAG, backend API development, and one major cloud AI platform.

Technical Toolkit

We value demonstrated depth and the ability to learn adjacent technologies—not superficial familiarity with every tool listed below.

• Agents & orchestration. LangGraph, LangChain, Google ADK, Anthropic Agent SDK, OpenAI SDK, CrewAI, AutoGen, or Semantic Kernel; agent orchestration, tools, memory, context management, observability, and evaluation.
• Retrieval systems. Embeddings, hybrid and semantic search, retrieval optimization, reranking, grounding, hallucination mitigation, and RAG evaluation; Pinecone, FAISS, Weaviate, ChromaDB, or similar vector stores.
• Backend & data. Python, FastAPI and/or Flask, REST APIs, microservices, clean architecture, SQL, and enterprise data platforms such as PostgreSQL, MySQL, or Snowflake.
• ML frameworks. PyTorch, TensorFlow, Scikit-learn, and Hugging Face Transformers; model training, fine-tuning, inference optimization, and NLP or computer vision applications.
• Cloud & AI platforms. AWS (Bedrock, AgentCore, SageMaker), Google Cloud (Gemini Enterprise, Vertex AI), or Microsoft Azure (Azure AI Foundry, Azure OpenAI); cloud-native deployment, monitoring, security, and scalability.
• DevOps & MLOps. Docker, Kubernetes, GitHub Actions or similar CI/CD tooling, MLflow, Kubeflow, SageMaker or comparable MLOps platforms; production model lifecycle and observability.
• AI-first engineering. GitHub Copilot, Cursor, OpenAI Codex, Claude Code, or Windsurf; practical use of AI-assisted engineering to improve development velocity, quality, and SDLC automation.

Preferred Qualifications

• Experience delivering customer-facing AI solutions in consulting or across multiple business domains, including healthcare, insurance, banking, retail, manufacturing, or financial services.
• Publications in leading AI, ML, NLP, computer vision, or data science journals and conferences; open-source contributions or participation in research communities.
• AWS, Google Cloud, Microsoft Azure, or AI-platform certifications and experience with enterprise security, governance, or Responsible AI practices.
• Demonstrated ability to evaluate emerging technologies, redesign workflows, and deliver measurable improvements in efficiency, quality, cost, or customer outcomes.

What We Look For

Strong problem-solving and engineering judgment; curiosity and continuous learning; clear communication and stakeholder management; and the ownership to deliver secure, scalable, production-ready AI solutions. You can balance speed with rigor, collaborate across functions, and explain both what works and what does not.

Skills

Python, Generative AI, FastAPI, Machine Learning