4Bell Technology

Job Description

Role:

As an AI/ML Engineer, you will build AI and agent features for an enterprise-grade SaaS platform in the data privacy, data governance and AI governance space. Your work will cover agent-based workflows, AI asset discovery and AI guardrails that detect and enforce controls for PII, toxicity and prompt injection, as well as new AI features taken from prototype to production.

Key Responsibilities:

·       Build agent-based features using LLMs and frameworks such as LangChain and LangGraph, including tool calling, orchestration and memory.

·       Develop AI asset discovery for models, agents, prompts and AI services across cloud and AI platforms.

·       Implement guardrail signals and enforcement, including PII detection and redaction, toxicity and content moderation, and prompt-injection and jailbreak detection.

·       Build RAG pipelines, prompt templates and evaluation harnesses.

·       Evaluate and benchmark models and guardrail classifiers for accuracy, latency and cost.

·       Take new AI features from prototype to production for new product initiatives.

·       Integrate AI services with backend microservices through well-defined APIs.

·       Add observability, logging and monitoring to AI components.

·       Work closely with the AI Architect on design and technical decisions.

What You’ll Bring:

·       Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field, or equivalent experience.

·       5–8+ years of software engineering experience, including at least 2–3 years building ML, GenAI or LLM applications in production.

·       Strong programming skills in Python.

·       Hands-on experience building agentic AI or LLM-powered features.

Mandatory Skills:

·       Strong proficiency in Python, with experience in FastAPI or Flask.

·       Experience in LLM application development: prompt engineering, RAG, embeddings and vector databases (e.g., pgvector, OpenSearch, Pinecone, Azure AI Search).

·       Hands-on experience with agent frameworks such as LangChain, LangGraph or Semantic Kernel, and familiarity with the Model Context Protocol (MCP).

·       Experience with AWS Bedrock/SageMaker and Azure OpenAI/Azure AI Foundry.

·       Experience implementing AI safety controls, such as PII detection (e.g., Presidio), content moderation and prompt-injection defense (e.g., NeMo Guardrails, Guardrails AI, Llama Guard).

·       Good understanding of ML and NLP fundamentals, including Hugging Face Transformers.

·       Experience with LLM evaluation frameworks and methods.

·       Familiarity with Git, Docker and CI/CD pipelines.

Good to Have:

·       Working knowledge of Java and Spring Boot.

·       Knowledge of AI governance frameworks such as NIST AI RMF and the EU AI Act.

·       Experience with MLOps practices and model fine-tuning.