Senior AI Engineer

Role: Senior AI Engineer(24147-1)

Duration: 6-12+ Months Contract

Location: Remote

Mandatory Skill :Agentic AI; Generative AI; LLMs; LangChain

Must Have Skill 1 Agentic AI / LLM orchestration with hands-on LangChain, LangGraph, LlamaIndex, Semantic Kernel, AutoGen, or Amazon Bedrock Agents

Must Have Skill 2 Generative AI / LLM application development, with minimum 2 years developing LLM/GenAI applications

Must Have Skill 3 RAG including embeddings, indexing strategies, vector retrieval, and retrieval-quality optimization

Must Have Skill 4 Python 3.11+, including type annotations, automated testing, asynchronous programming, APIs/FastAPI

Must Have Skill 5 LLM Evaluation, including golden datasets, metrics, non-deterministic output evaluation, and LLM-as-judge/evaluation frameworks

Must Have Skill 6 Enterprise production AI deployment & post-production support including incident response, regression management, latency/cost management, and live production changes

Must Have Skill 7 AI Observability & Cloud Services including tracing, structured logging, quality/cost monitoring, REST APIs/data pipelines/event-driven services

Must Have Skill 8 Enterprise Security & Compliance, including security/compliance reviews, access control, data handling, and auditability

Domain Experience Regulated industry strongly preferred, particularly Financial Services. Insurance and Healthcare are also acceptable preferred backgrounds.

Experience Minimum 5 years software engineering + minimum 2 years LLM/GenAI application development

We are seeking a Senior AI Engineer to design, build, deploy, and operate enterprise-grade Large Language Model (LLM) applications supporting business processes within a regulated financial services environment.

The ideal candidate will have hands-on experience taking Generative AI solutions from design through production deployment and providing ongoing production support. This role requires deep expertise in LLM orchestration, RAG, AI evaluation, Python, cloud services, observability, and production-grade AI engineering.

____

Key Responsibilities

• Design and implement multi-step LLM pipelines, including:

o LLM orchestration

o Structured output generation

o Tool/function calling

o Error handling, retry, and fallback logic

• Develop and maintain Retrieval-Augmented Generation (RAG) solutions, including:

o Indexing strategies

o Embedding model selection

o Retrieval quality optimization

• Build and maintain LLM evaluation frameworks and golden datasets to measure accuracy, consistency, and regression risk before production releases.
• Establish and enforce quality gates for model, prompt, and configuration changes.
• Implement observability for non-deterministic AI applications, including:

o Tracing

o Structured logging

o Quality monitoring

o Cost monitoring

• Manage cost and latency performance of production AI workloads against defined budgets.
• Partner with Product, Business, Risk, Compliance, and Security teams to translate business requirements into technical solutions and support review/approval processes.
• Participate in code reviews, architectural decisions, and technical documentation.
• Provide ongoing production support for deployed AI applications, including incident response, root-cause analysis, regression management, and remediation.
• Mentor junior engineers and contribute to engineering standards and best practices.

____

Required Qualifications

• Bachelor's degree in Computer Science, Engineering, or a related technical field, or equivalent practical experience.
• Minimum 5 years of professional software engineering experience.
• Minimum 2 years of experience developing LLM/Generative AI applications.
• Demonstrated experience deploying at least one enterprise-grade AI application into production and supporting it after launch.

Enterprise Production AI Experience Must Include

• AI solution serving real users or business processes, rather than only a proof of concept or internal pilot.
• Accountability for non-functional requirements such as:

o Availability

o Latency

o Cost management

o Access control

o Data handling

o Auditability

• Experience completing security and compliance reviews within a regulated or enterprise environment.
• Post-production ownership involving:

o Incident response

o Regression management

o Changes deployed against live traffic

Technical Requirements

• Hands-on experience with at least one LLM agent/orchestration framework, such as:

o LangGraph

o LangChain

o LlamaIndex

o Semantic Kernel

o AutoGen

o Amazon Bedrock Agents

• Strong experience with LLM evaluation methodologies, including:

o Dataset/golden dataset curation

o Metric selection

o Evaluation of non-deterministic outputs

o Appropriate use of LLM-as-judge techniques

o Established evaluation frameworks or custom evaluation harnesses

• Strong proficiency in Python 3.11+, including:

o Type annotations

o Automated testing

o Asynchronous programming

• Experience building and operating cloud-based services such as:

o REST APIs

o Data pipelines

o Event-driven workers

• Experience working within codebases governed by automated quality gates and formal code reviews.

____

Preferred Qualifications

• Experience with Document AI / Intelligent Document Processing, including:

o OCR

o Structured field extraction

o Scanned documents

o Photographed documents

o Handwritten source material

• Experience designing human-in-the-loop workflows, including:

o Confidence scoring

o Selective automation

o Escalation and review processes

• Experience implementing business rules or policies as versioned and reviewable artifacts.
• Experience with event-driven architecture, particularly Apache Kafka.
• Experience with Kubernetes, Helm, and CI/CD pipeline ownership.
• Experience with Infrastructure as Code, particularly Terraform.
• Experience with:

o Model selection

o Model benchmarking

o Fine-tuning

o Distillation

o Cost and quality optimization

• Prior experience within financial services, insurance, healthcare, or another regulated industry.

____

Technical Environment

AI / GenAI

• LLMs
• LLM orchestration frameworks
• RAG
• Agentic AI
• LLM evaluation and observability tooling

Programming / Application Development

• Python 3.11+
• FastAPI
• Pydantic

Data / Database

• PostgreSQL with vector extensions
• Aurora PostgreSQL

Cloud / AI Platform

• AgentCore

Infrastructure / DevOps

• Kubernetes
• Helm
• Terraform
• CI/CD

Event-Driven Architecture

• Apache Kafka

Observability

• Datadog
• OpenTelemetry

Key Must-Have Skills for Screening

Python 3.11+ | LLM/GenAI | RAG | LangGraph/LangChain/LlamaIndex/Semantic Kernel/AutoGen/Bedrock Agents | LLM Evaluation | Production AI Deployment & Support | Cloud Services/APIs | AI Observability | Enterprise Security/Compliance

Most important screening point: The candidate should have actually deployed an enterprise-grade GenAI/LLM application to production and supported it post-launch. POC-only GenAI experience would not meet the stated requirement.