AI Engineer

Shakti Solutions; Atlanta (GA); On-site; USD 150,000 - 230,000; Full time

Job summary

Shakti Solutions is seeking a senior AI/ML engineer to design and deploy agentic systems for regulated processes. You will build multi-step reasoning, planning, and tool use workflows using LangGraph/LangChain, with emphasis on reliability and auditable decision making.

You will implement observability with Azure Monitor, manage memory and context, and integrate with enterprise APIs. Remote collaboration with a multidisciplinary team is expected.

Qualifications

• Proven production experience building agentic systems, not just exploration.
• Hands‑on experience with LangGraph/LangChain or equivalent orchestration.
• Experience building end‑to‑end RAG systems: indexing, retrieval, reranking, grounding.

Responsibilities

• Design and build agentic systems for multi-step reasoning, planning, tool use, and workflow execution in regulated processes.
• Build stateful workflows with LangGraph/LangChain — branching, retries, self‑correction, human‑in‑the‑loop checkpoints.
• Engineer for reliability: error recovery, planning under uncertainty, robust handling of failed tool calls.
• Build auditable, policy‑grounded reasoning for high‑stakes decisions (e.g., prior authorization, claims review).
• Manage conversational state, persistent memory, and context assembly; apply MCP‑style tool/context interfaces.

Skills

Python

LangChain

Agentic systems

CI/CD

Observability

Azure

Education

Bachelor's degree in CS or related field

Tools

LangGraph

Azure AKS/ARO

Key Vault

Redis

Kafka

Job description

Full-time or contract W-2 with Shakti Solutions - Visa-independent resources only.

Location: Chicago, IL / Atlanta, GA / NYC

Key Role Activities

• Design and build agentic systems for multi-step reasoning, planning, tool use, and workflow execution in regulated processes.
• Build stateful workflows with LangGraph/LangChain — branching, retries, self‑correction, human‑in‑the‑loop checkpoints.
• Engineer for reliability: error recovery, planning under uncertainty, robust handling of failed tool calls.
• Build auditable, policy‑grounded reasoning for high‑stakes decisions (e.g., prior authorization, claims review).
• Manage conversational state, persistent memory, and context assembly; apply MCP‑style tool/context interfaces.
• Implement observability and tracing (Azure Monitor/Application Insights) for prompts, tool calls, and agent behavior.
• Apply guardrails to reduce hallucinations and unsafe actions; evaluate agents at the task and trajectory level.
• Support PHI/HIPAA‑aware data handling and human‑oversight/escalation for regulated decisions.
• Integrate agents with enterprise systems and APIs (e.g., MuleSoft as an integration layer).
• Deploy and operate on Azure — AKS/ARO, Key Vault, Redis, Kafka, Istio, networking.
• Deliver production‑quality code with strong testing, CI/CD, and documentation practices.

Required Qualifications

• Demonstrated production experience building agentic systems, not just exploration.
• Hands‑on experience with LangGraph/LangChain or equivalent orchestration.
• Experience building end‑to‑end RAG systems: indexing, retrieval, reranking, grounding, evaluation.
• Solid understanding of context/memory management and retrieval‑driven context assembly.
• Practical understanding of LLM limitations, hallucination risks, and evaluation methods.
• Experience debugging agent behavior at the trajectory/task level.
• Strong Python skills: testing, CI/CD, version control, API integration, production observability.
• Hands‑on experience with at least one frontier model platform (Anthropic, Google, OpenAI).
• Working knowledge of Azure infrastructure — AKS/ARO, Key Vault, Redis, Kafka, Istio, networking.
• Clear communication and problem solving skills, ability to meet deadlines, plan work and pivot as needed, and work with a multidisciplinary, diverse team.