Compunnel, Inc.
We are seeking an experienced Agentic AI Engineer to design and implement GenAI agentic solutions that enhance reliability, reduce risk, and optimize cost in large-scale production environments.
This role focuses on building intelligent agents capable of diagnosing, reasoning, and executing actions in runtime ecosystems to support production operations and improve productivity.
Key Responsibilities
Agentic AI System Development: Design and implement tool-calling agents that integrate retrieval, structured reasoning, and secure action execution (e.g., function calling, policy enforcement) using MCP protocol.
Engineer safety guardrails and enforce least-privilege access.
LLM Productionization: Build evaluation frameworks for open-source and foundational LLMs.
Implement retrieval pipelines, prompt synthesis, response validation, and self-correction loops tailored to production operations.
Runtime Ecosystem Integration: Connect agents to observability, incident management, and deployment systems to enable automated diagnostics, runbook execution, remediation, and post-incident summarization with traceability.
User Collaboration: Partner with production engineers and application teams to translate operational challenges into agentic AI roadmaps.
Define objective functions linked to reliability, risk reduction, and cost efficiency.
Safety, Reliability & Governance: Build validator models, adversarial prompts, and policy checks. Enforce deterministic fallbacks, circuit breakers, and rollback strategies.
Instrument continuous evaluations for usefulness, correctness, and risk.
Performance Optimization: Improve cost and latency through prompt engineering, context management, caching, model routing, and distillation.
Use batching, streaming, and parallel tool-calls to meet service-level objectives under real-world load.
RAG Pipeline Development: Curate domain knowledge, build data-quality validation frameworks, and establish feedback loops to maintain knowledge freshness.
Engineering Excellence: Lead design reviews, promote rigorous experimentation, and mentor peers on agent architectures, evaluation methodologies, and safe deployment practices.
Required Qualifications
5+ years of software development experience in Python, C/C++, Go, or Java (Python preferred).
3+ years of experience designing and launching production ML systems, including model deployment, evaluation, monitoring, and fine-tuning workflows.
Practical experience with LLMs: API integration, prompt engineering, fine-tuning, and building RAG-based applications with tool-using agents.
Familiarity with commercial and open-source LLMs (e.g., OpenAI, Gemini, Llama, Qwen, Claude).
Strong foundation in applied statistics, machine learning concepts, algorithms, and data structures.
Excellent problem-solving and communication skills with a focus on measurable business impact.
Preferred Qualifications
Experience with cloud infrastructure (preferably AWS), including ECS/EKS, Lambda, S3, DynamoDB, Redshift, Step Functions, SageMaker, and infrastructure-as-code tools like Terraform or CloudFormation.
Participation in platform modernization and performance engineering initiatives.
#J-18808-Ljbffr
This role focuses on building intelligent agents capable of diagnosing, reasoning, and executing actions in runtime ecosystems to support production operations and improve productivity.
Key Responsibilities
Agentic AI System Development: Design and implement tool-calling agents that integrate retrieval, structured reasoning, and secure action execution (e.g., function calling, policy enforcement) using MCP protocol.
Engineer safety guardrails and enforce least-privilege access.
LLM Productionization: Build evaluation frameworks for open-source and foundational LLMs.
Implement retrieval pipelines, prompt synthesis, response validation, and self-correction loops tailored to production operations.
Runtime Ecosystem Integration: Connect agents to observability, incident management, and deployment systems to enable automated diagnostics, runbook execution, remediation, and post-incident summarization with traceability.
User Collaboration: Partner with production engineers and application teams to translate operational challenges into agentic AI roadmaps.
Define objective functions linked to reliability, risk reduction, and cost efficiency.
Safety, Reliability & Governance: Build validator models, adversarial prompts, and policy checks. Enforce deterministic fallbacks, circuit breakers, and rollback strategies.
Instrument continuous evaluations for usefulness, correctness, and risk.
Performance Optimization: Improve cost and latency through prompt engineering, context management, caching, model routing, and distillation.
Use batching, streaming, and parallel tool-calls to meet service-level objectives under real-world load.
RAG Pipeline Development: Curate domain knowledge, build data-quality validation frameworks, and establish feedback loops to maintain knowledge freshness.
Engineering Excellence: Lead design reviews, promote rigorous experimentation, and mentor peers on agent architectures, evaluation methodologies, and safe deployment practices.
Required Qualifications
5+ years of software development experience in Python, C/C++, Go, or Java (Python preferred).
3+ years of experience designing and launching production ML systems, including model deployment, evaluation, monitoring, and fine-tuning workflows.
Practical experience with LLMs: API integration, prompt engineering, fine-tuning, and building RAG-based applications with tool-using agents.
Familiarity with commercial and open-source LLMs (e.g., OpenAI, Gemini, Llama, Qwen, Claude).
Strong foundation in applied statistics, machine learning concepts, algorithms, and data structures.
Excellent problem-solving and communication skills with a focus on measurable business impact.
Preferred Qualifications
Experience with cloud infrastructure (preferably AWS), including ECS/EKS, Lambda, S3, DynamoDB, Redshift, Step Functions, SageMaker, and infrastructure-as-code tools like Terraform or CloudFormation.
Participation in platform modernization and performance engineering initiatives.
#J-18808-Ljbffr