AI Engineer - Agentic & GenAI Systems

Role Summary

Design, build, and operate production-grade agenticand GenAI systems—end to end. You’ll ship services (not just notebooks): robustAPIs, reusable components, and secure pipelines that connect LLMs, tools,knowledge, and enterprise systems. You’ll pair strong software engineering withmodern AI practices (RAG, agent orchestration, policy chains, evals) to delivermeasurable business outcomes at scale.

Responsibilities

Agent & Application EngineeringS

• Design multi-agent systems (MAS) with planning, tool-use, and delegation (e.g., LangGraph/Semantic Kernel); expose them via REST/gRPC APIs (FastAPI/Express/Java/Go).
• Implement tool adapters (SQL, search, document stores, web calls, code exec) with strict type contracts and safe sandboxes.
• Build model gateway integrations (OpenAI/Azure OpenAI/Bedrock/Vertex; self-hosted vLLM/TGI) with routing, rate-limits, retries, and fallback chains.
• Orchestrated workflows with LangChain or LlamaIndex — planning, tool calling, structured extraction, multi-step state
• Typed contracts on every model output using Pydantic schemas, so downstream code never parses free text
• Guardrails and policy enforcement at the boundary, with deterministic rules taking precedence over the model

Retrieval, Data & Knowledge

• Stand up RAG services: chunking, enrichment, embeddings, indexing, hybrid/vector search (pgvector/Pinecone/Weaviate; OpenSearch/Azure AI Search).
• Implement ingestion pipelines (Airflow/Prefect/Celery/Ray) for docs, tickets, chat, and ERP/CRM data; handle PII redaction and metadata governance.
• Optimize retrieval quality (chunking strategies, re-rankers, query rewriting) with offline/online evaluation and A/B tests.

Quality, Testing & Evaluation

• Treat prompts and graphs as code: version, diff, and test them (unit tests for prompts/tools; golden sets; regression suites).
• Build evaluation harnesses (latency, cost, accuracy, toxicity, hallucination, guardrail hit-rates); wire into CI.
• Add drift detection for conversational systems; implement safe shutdown and auto-rollback.

Platform & Operations

• Package services as containers; deploy to /AKS/EKS and CI/CD with Helm/Argo CD; configure autoscaling, HPA/VPA, and resource quotas.
• Implement policy chains and guardrails (OPA/Gatekeeper for policy, Presidio for PII, Trivy for image scanning).
• Instrument deep observability: tracing (OpenTelemetry), metrics (Prometheus), logs (ELK/OPENSEARCH), cost meters per request/model.

Security & Compliance

• Manage secrets (HashiCorp Vault/KMS), signed images, SBOMs; enforce least-privilege IAM.
• Build tenant isolation and data residency controls; implement red/blue team prompts and jailbreak defenses.

Integration & Enterprise Workflows

• Ship connectors and events for SAP/CRM/ITSM and Kafka topics; design idempotent, retry-safe processors.
• Automate business workflows with pro-code services first; expose low-code surfaces only where appropriate.

Collaboration & Leadership

• Partner with Product, Data, and Platform teams to define SLAs/SLOs and success metrics.
• Mentor engineers on “AI as software” practices; run design reviews and postmortems.

Minimum Qualifications

• 6+ years in software engineering (prod services, not just prototypes), including 1+ year leading small projects.
• Strong in one systems language (Python/TypeScript/Go/Java) and comfortable in a second.
• Hands-on with containers, Kubernetes, CI/CD (GitHub Actions/GitLab/Jenkins), IaC (Terraform), and cloud (Azure/AWS/GCP).
• Practical LLM experience: building RAG/agent apps, prompt design, tool-use, and safety patterns.
• Data skills: designing schemas, batch/stream pipelines, and search indexes; proficiency with SQL and one vector DB.
• Testing mindset: unit/integration tests, load tests, golden datasets for LLM evals.
• Security basics: secrets, policies, scanning, and least-privilege IAM.

Preferred Qualifications

• Agent orchestration (LangGraph, Semantic Kernel) and distributed compute (Ray) in production.
• Search/retrieval tuning (BM25 + vector hybrid, re-ranking, query planning).
• Observability at scale with OpenTelemetry; cost/perf optimization across model/router layers.
• Experience in regulated or high-throughput domains (e.g., telco, finance, healthcare); multi-tenant and data-residency patterns.
• Domain integrations (SAP/CRM/ITSM), event-driven architectures (Kafka/Debezium), and policy enforcement (OPA/Gatekeeper).
• Familiarity with Order to Cash Process flows in a Telco Enterprise environment is a plus

Tech Stack (Illustrative)

• Languages: Python, TypeScript/Node.js (plus Go/Java bonus)
• Frameworks: FastAPI/Express, LangGraph/Semantic Kernel, Ray/Celery, Airflow/Prefect
• Storage/Search: Postgres, Redis, S3/Blob; pgvector/Pinecone/Weaviate; OpenSearch/Azure AI Search
• LLM Runtime: OpenAI/Azure OpenAI/Bedrock/Vertex; vLLM/TGI; inference routers/gateways
• Platform: Docker, Kubernetes, Helm, Argo CD, Terraform, Vault, Istio
• Observability: OpenTelemetry, Prometheus/Grafana, ELK/OpenSearch
• Quality & Safety: pytest/Jest, prompt/unit test harnesses, guardrails, Presidio, Trivy