AI Engineer

INVOKE; United States; On-site; USD 140,000 - 190,000; Full time

Job summary

INVOKE is a consulting and Innovation firm focused on intelligent automation. This role supports U.S. public sector clients, building LLM-powered automations and AI services that integrate with client systems.

You will design retrieval-augmented generation pipelines, implement tool calling, and deploy solutions using FastAPI or Node.js, Docker/Kubernetes, and serverless or containerized architectures. Collaboration with product, delivery, and client stakeholders is essential.

Qualifications

• Bachelor’s or Master’s degree in CS, Engineering, or related field.
• 2–3 years of software engineering experience with Python/Java/Node.js/C# or similar.
• 1–2 years hands-on with LLMs: prompt engineering, RAG, agents, and structured outputs.
• Familiar with LLM SDKs/frameworks and vector databases.
• Experience with ingestion/embeddings for unstructured data and document stores.
• Basic DevOps/MLOps skills: Docker, Kubernetes or serverless, CI/CD, IAM.
• Exposure to AWS/Azure/GCP and safe coding practices; strong communication.

Responsibilities

• Build LLM-powered automations, chat/voice assistants, and intelligent agents.
• Design and deploy retrieval-augmented generation pipelines with document ingestion and embeddings.
• Implement tool/function calling and multi-step agent workflows with human-in-the-loop when needed.
• Package solutions as services (FastAPI/Node.js) with tests, observability, and CI/CD; deploy to cloud.
• Instrument and tune LLM solutions, manage tracing, latency, and cost budgets; run evaluations.
• Implement guardrails, PII redaction, schema validation, and model routing across providers.
• Collaborate with product, delivery, and client stakeholders; run POCs and scale prototypes.
• Participate in code reviews and help improve internal templates, tooling, and docs.

Skills

Python

Java

Node.js

LLM experience

Prompt engineering

Vector databases

OpenAI/LLM frameworks

Cloud platforms

DevOps basics

Communication

Education

Bachelor’s or Master’s degree in CS/Engineering

Tools

LangChain

LlamaIndex

Pinecone

FAISS

OpenAI/Azure OpenAI

Docker

Kubernetes

Job description

Invoke is a consulting and Innovation firm focused on delivering Intelligent automation solutions that solve real operational challenges. We partner with organizations across industries, including the public sector, to design and implement scalable RPA and AI-driven workflows that improve efficiency, accuracy, and reliability.

Our teams work hands-on with clients to automate complex processes across a range of systems, from modern applications to legacy environments. A significant portion of our work is within the public sector, where we help government organizations streamline high-volume processes, improve data quality, and modernize critical operations through automation.

We are a fast growing organization, always looking for new members to our family. INVOKE provides the technology, implementation, and lifecycle expertise to solve business challenges through the lens of intelligent automation technologies like Robotic Process Automation and Artificial Intelligence.

As this role supports public sector clients, U.S. citizenship is required.

What You’ll Be Doing

• Build LLM-powered automations, chat/voice assistants, and intelligent agents that integrate seamlessly with client systems.
• Design and deploy retrieval-augmented generation (RAG) pipelines, including document ingestion, chunking, embeddings, vector search, and grounding for accurate, auditable responses.
• Implement tool/function calling and multi-step agent workflows to perform actions (draft → review → execute → verify), incorporating human-in-the-loop processes where necessary.
• Package solutions as reliable services (e.g., FastAPI or Node.js) with tests, observability, and CI/CD pipelines; deploy to the cloud using serverless or containerized architectures.
• Instrument, evaluate, and tune LLM solutions—manage tracing, latency and cost budgets, prompt/version control, A/B tests, and golden-set evaluations to reduce hallucinations and improve output quality.
• Implement guardrails and safety mechanisms, including content filters, PII redaction, schema/JSON validation, fallbacks, and model routing across providers.
• Collaborate with product, delivery, and client stakeholders to scope use cases, run quick proofs of concept (POCs), and scale successful prototypes into production-ready systems.
• Participate in code reviews and contribute to improving internal templates, tooling, and documentation to enhance reliability and development speed.
• Occasionally support hiring initiatives through interviews or technical assessments.

Qualifications

• Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field.
• 2–3 years of experience in software engineering using one or more programming languages: Python, Java, Node.js, C#, or similar.
• 1–2 years of hands-on experience building with LLMs—prompt engineering, RAG, agents, tool/function calling, structured outputs, and streaming.
• Proficiency with LLM SDKs and frameworks (e.g., OpenAI/Azure OpenAI, Anthropic, Google, LangChain, LlamaIndex) and vector databases (e.g., Pinecone, Weaviate, pgvector/Postgres, FAISS).
• Experience preparing unstructured data (PDFs, HTML, emails, tickets) and developing robust ingestion/embedding pipelines and document stores (e.g., S3, GCS).
• Familiarity with evaluation and observability tools (e.g., LangSmith, RAGAS/DeepEval, OpenTelemetry, logging) and experience writing automated tests for LLM workflows.
• Basic DevOps/MLOps skills: Docker, Kubernetes or serverless (Lambda, Cloud Run), CI/CD (GitHub Actions), and secrets/IAM best practices.
• Exposure to cloud platforms (AWS, Azure, GCP) and related services (API Gateways, managed databases/queues; Bedrock or Azure OpenAI experience is a plus).
• Strong problem-solving and communication skills; ability to translate business workflows into practical automations and clearly explain trade-offs to non-technical stakeholders.

Nice to Have

• Experience with fine-tuning or LoRA adapters for domain-specific tasks; knowledge of prompt caching and cost optimization strategies.
• Experience integrating with enterprise applications and knowledge bases, and developing lightweight admin UIs (React, Next.js) for internal tools.
• Strong security mindset, including familiarity with OAuth/JWT, least-privilege access, PII handling, and compliance best practices