Additional Job Description

Minimum Degree Required: B.E/B.Tech

Degree Preferred: B.E/B.Tech

Minimum Years of Experience: 5-8 Years

- 4+ year of experience working with AI solutions

Required Technical Knowledge / Skills

- Design and implement Generative AI solutions leveraging large language models, Retrieval-Augmented Generation architecture patterns, semantic search, knowledge graphs, and enterprise knowledge integration.
- Implement Agentic AI solutions, including autonomous AI workflows, multi-agent orchestration, agentic frameworks, tool integration, guardrails, and human-in-the-loop controls for enterprise use cases.
- Establish prompt engineering, prompt management, evaluation, versioning, reuse, and lifecycle practices for scalable Generative AI delivery.
- Drive adoption of DevOps and MLOps practices, including CI/CD, infrastructure as code, automated testing, model deployment automation, monitoring, alerting, and release governance.
- Implement responsible AI and AI governance practices, including AI risk assessments, model explainability, transparency, validation, monitoring, and regulatory readiness for AI systems.
- Ability to support technical configuration activities including group and security configuration, database rules, validations, calculation logic, and reference data definition

Strongly preferred:

- Python: Strong hands-on Python experience, including async (asyncio/anyio), type hints, and Pydantic v2. Experience building FastAPI services with versioned API contracts.
- LLM application development: Hands-on experience building LLM applications with Pydantic AI or similar frameworks. Includes typed,



schema validated outputs and handling malformed model responses through retries, repair, and fallbacks.
- Workflow orchestration: Experience building multi-step LLM pipelines with LangGraph.
- Document AI: Extracting data from PDFs, scans, images, and Word files using OCR (Tesseract, PaddleOCR) and PDF tooling (PyMuPDF, pypdf). Includes deciding when to use native text and when to use OCR.
- LLM evaluation: Building ground-truth datasets, field-level scoring, baseline comparisons, and CI regression gates, so prompt and model changes are measured, not guessed.
- Prompt engineering: Designing, versioning, and tuning prompts for accuracy, token cost, and latency.
- LLM security: Mitigating prompt injection, handling untrusted input safely, and returning safe error responses.
- Production delivery: A track record of shipping GenAI systems to production. Sound judgment on trade-offs between quality, latency, cost, and scalability, including batching, retries, confidence thresholds, and routing to human review.

Preferred Skills:

- Docker and Kubernetes (Helm, GitOps), plus CI/CD with GitHub Actions.
- Testing with pytest, including fixture-driven tests and mocked LLM clients.
- Integrating with enterprise workflow platforms such as Appian, or other downstream systems with strict API contracts.
- Observability for LLM calls, including tracing and token, and cost tracking.

Certifications (optional):

- A cloud AI/ML certification, such as Microsoft Azure AI Engineer Associate, AWS Certified Machine Learning Engineer Associate, or Google Cloud Skilled Machine Learning Engineer
- Certified Kubernetes Application Developer (CKAD)
- Databricks Certified Generative AI Engineer Associate, or a similar vendor GenAI certification

📌 Python AI Engineer (Hyderabad)
🏢 PwC
📍 Hyderabad