AI Platform Engineer

Direct Responsibilities

- Design, develop, and deploy autonomous and semi‑autonomous AI agents that automate CRM workflows (sales assistance, support automation, lead qualification) and orchestrate multi‑step processes across enterprise data. Implement permission‑aware tool adapters, bounded autonomy, guard‑rails, escalation mechanisms and full audit‑logging to satisfy AML and security compliance

- Build production‑grade Retrieval‑Augmented Generation (RAG) pipelines that ingest structured and unstructured data, perform chunking, embedding, hybrid search, semantic retrieval, metadata filtering and re‑ranking using vector databases (pgvector, Qdrant, Pinecone, Weaviate). Create reusable LLM abstraction layers that support multiple model providers and open‑weight models, with model routing, fall‑backs, rate‑limiting, cost‑tracking and token‑consumption optimization (caching, context‑window management)

- Design and implement robust state & memory management for autonomous agents—handle checkpointing, persistent context storage, efficient serialization/retrieval, and scalable memory‑access patterns to support long‑running, multi‑turn interactions while maintaining consistency and compliance.

- Develop AI‑powered CRM capabilities such as customer profiling, lead qualification, meeting preparation, interaction summarization, follow‑up generation and next‑best‑action recommendation. Build natural‑language interfaces that let users query customer information, generate insights and trigger workflows conversationally. Enable agents to retrieve context, update CRM records and execute authorized actions while integrating LLM reasoning with business‑rule engines and existing CRM processes

- Engineer scalable backend services and APIs (Python, FastAPI, async) that expose agent functionality, integrate with CRM systems, databases, email/calendar services, document repositories and external APIs. Implement reliable execution patterns – background jobs, message queues, retries, timeouts, idempotency and fault tolerance – and construct event‑driven architectures with Kafka or RabbitMQ.

Design and operate AI‑infrastructure and LLMOps pipelines: containerize services (Docker), orchestrate on‑prem or cloud‑native Kubernetes clusters (Helm, Argo CD GitOps), manage CI/CD (GitHub Actions, Jenkins), and implement observability (Prometheus, Grafana, ELK) together with model‑lifecycle tooling (MLflow, experiment tracking, model registry). Optimize latency, throughput, concurrency, resource utilization and inference cost; enforce access controls, tenant isolation, secrets management and audit trails.

- Provide technical leadership across product, engineering and compliance teams: own end‑to‑end architecture, prototype, productionise and maintain AI solutions; develop reusable frameworks and standards for agentic AI; conduct code and architecture reviews; mentor junior engineers; and continuously evaluate emerging agentic‑AI frameworks, models and deployment technologies for product relevance.

Contributing Responsibilities

· Proficiency in tools (e.g., JIRA, Confluence, MS Visio)

· Familiarity with software development methodologies (Agile, Scrum, Waterfall).

· Familiarity with Oracle databases, Understanding schema, SQL

· Strong analytical, communication, and interpersonal skills

· Ability to work effectively with cross-functional teams and manage multiple priorities.

· Quick to learn and adapt

· Good Team player 

· Someone who could have long-term commitment with the team and organization