Role Overview
deeply hands-on AI professional
This is a builder role. Screening will emphasize demonstrable, hands-on delivery of production Generative AI systems not slideware or PoCs.
Key Responsibilities
AI Solution Design & Delivery
- Design and develop end-to-end Generative AI applications for enterprise use cases such as document intelligence, knowledge assistants, and workflow automation.
- Build production-grade RAG pipelines with strong grounding, evaluation, and observability.
- Develop agentic AI workflows and multi-agent systems for complex reasoning and orchestration tasks.
- Build multimodal AI solutions capable of processing text, PDFs, tables, images, and scanned documents.
Architecture & Engineering
- Own end-to-end solution architecture: data ingestion, model layer, orchestration, APIs, evaluation, and deployment.
- Build scalable backend APIs and microservices in
Python (FastAPI)
with clean, testable, well-documented code. - Deploy and maintain AI solutions on cloud-native infrastructure (GCP preferred; AWS/Azure acceptable) using containers and CI/CD.
- Implement MLOps/LLMOps practices versioning, monitoring, cost governance, and rollback safety.
Evaluation, Governance & Compliance
- Design AI evaluation frameworks covering accuracy, groundedness, faithfulness, hallucination control, and bias.
- Instrument Human-in-the-Loop (HITL) gates and audit-ready logs suitable for regulated environments.
- Align outputs with data integrity, GxP, and enterprise security requirements.
Collaboration, Ownership & Innovation
- Partner with the Principal Product Manager, Principal Data Scientist, business SMEs, and cloud partners to translate requirements into production-ready systems.
- Assimilate detailed project skill files, PRDs, and technical specs shared for different initiatives, and rapidly convert them into working solutions.
- Bring fresh ideas and challenge existing designs with data-backed recommendations.
- Mentor junior team members and interns; contribute to internal AI best practices.
Qualifications
Education
- Bachelor's or Master's degree in Computer Science, Data Science, Artificial Intelligence, or a related quantitative field from a reputed institute.
Experience
4 to 7 years
overall in AI Engineering / Data Science / ML.At least 2 to 3 years of hands-on, production-grade LLM / Generative AI development
building, deploying, and maintaining real systems (not prototypes).- Remaining experience in strong software engineering or applied ML delivery is acceptable.
Must-Have Skills (Hands-on, Production-grade)
Generative AI
- Hands-on experience with LLMs such as GPT, Gemini, or Anthropic models, and transformer-based architectures.
- Retrieval-Augmented Generation (RAG) chunking strategies, embeddings, vector stores, and re-ranking, deployed in production.
- Prompt engineering and prompt versioning discipline.
- LLM evaluation frameworks and metrics (groundedness, faithfulness, coverage, hallucination).
Programming & Engineering
Python (Advanced)
production code, testing, packaging, and performance tuning.- SQL analytical queries on modern warehouses.
FastAPI
(or Django) for backend service development.- Firebase (or equivalent), Git, REST APIs, Docker, CI/CD basics, and Linux.
Cloud & Delivery
- Production experience on at least one of GCP, AWS, or Azure; comfort with IAM, storage, compute, and managed AI services.
- Strong software engineering and system design skills, with proof of taking AI systems from prototype to production with defined SLAs.
Good-to-Have Skills
Advanced AI / GenAI
- Building agents on existing agent platforms (e.g., Vertex AI Agent Builder, Copilot Studio, LangGraph, CrewAI, or equivalent).
- Multi-agent orchestration for complex enterprise workflows.
- Multimodal AI (text + images + tables + scanned documents) and document AI (OCR, layout models, hierarchical extraction).
- Fine-tuning, distillation, or adapter-based customization of LLMs.
Enterprise Data & Knowledge Platforms
- Exposure to designing or contributing to enterprise data & knowledge platforms — canonical data models, semantic layers, or knowledge graphs that power multiple downstream AI applications.
- Familiarity with graph databases (Neo4j, AuraDB, or similar) and knowledge-graph patterns.
- Modern data stack experience: BigQuery, dbt, and workflow orchestrators such as Airflow or Cloud Composer.
