Job DetailsSectionDetails / Example ContentJob Title*Generative AI DeveloperJob ID Coding* MandatoryExperience Level*Min* 5yrsMax* 10yrsLocation*On-siteWork Type*Full-time

Section II - Job Evaluation Topics and Weight (Questions and evaluation are based on below)Topic of Evaluation / Skills* Mandatory / Non-mandatoryPercentage*
(Skills / Topics should sum up to 100%)
(Enter only multiples of 5)Python concepts and FrameworkMandatory30%Docker and MicroservicesMandatory5%PostgreSQL/Mongo DBMandatory5%LangChainMandatory15%LangGraphMandatory10%Google ADKMandatory10%Fast APIMandatory10%RAGMandatory5%Multi Agent architectureMandatory10%

Section III - Job Requirements and ResponsibilitiesSectionDetails / Example ContentJob Requirements*

- 2 to 6 Years of experience using Python.
- Strong proficiency in Python and experience with relevant libraries and frameworks (e.g., pandas, LangChain, LangGraph).
- Strong experience working with API calls (FAST API, Rest API)
- Proficiency with LLM integrations
- At least one project is live in production
- Ability to work efficiently within a fast-paced, deadline-driven environment, prioritize tasks, manage time effectively, and communicate challenges or issues proactively
- Good to have experience working using MCP servers

Key Responsibilities*Development & Implementation

- Implement Python modules for data ingestion, prompt orchestration, and basic tool-agent workflows.
- Build REST APIs/microservices (FastAPI/Flask) to connect LLMs with enterprise data sources.
- Assist in developing Retrieval-Augmented Generation (RAG) components: ingestion, embeddings, chunking, and retrieval.




- Configure and integrate LLMs (OpenAI, Gemini, Claude, Llama, Mistral) via APIs or SDKs.
- Write unit/integration tests; contribute to robust error handling and logging.
- Support feature deployment and troubleshoot issues in staging/production.

2. Data & Model Ops

- Prepare datasets, metadata, and embeddings using vector databases (FAISS, Chroma, Pinecone, Weaviate).
- Participate in evaluating model outputs; assist with prompt optimization and context management.
- Help maintain experiment tracking and model/version metadata using MLOps tools (e.g., MLflow).
- Contribute to metrics dashboards (latency, token usage, cost, accuracy).

3. Cloud Integration

- Assist in deploying workloads to one hyperscaler (GCP, Azure, or AWS) under guidance:
- GCP: Vertex AI, Cloud Run basics
- Azure: Azure OpenAI, Cognitive Search fundamentals
- AWS: Bedrock basics, Lambda/API Gateway wiring

- Follow IAM/security best practices and cost-awareness guardrails.

4. Collaboration & Quality

- Work in an agile squad with senior engineers, architects, and product owners.
- Participate in code reviews and adhere to coding standards, documentation, and secure development practices.
- Support demo builds, POCs, and reusable component libraries.




- Communicate clearly on progress, blockers, and risks.

5. Continuous Learning & Innovation

- Stay current with GenAI frameworks (LangChain, LangGraph, Google ADK), embeddings, and evaluation methods.
- Experiment with PEFT/LoRA, prompt engineering techniques, and basic fine-tuning workflows.
- Contribute to internal playbooks, examples, and knowledge-sharing sessions.

Good to HaveKnowledge on A2A, Async Programming and Cloud Computing
Section IV - Job Qualifications & SkillsSectionDetails / Example ContentDomainNASoft Skills- Excellent communication
- Team collaboration
- Documentation and knowledge sharingEducation RequirementsBachelor's/masters in computer science or equivalent (can be marked optional or adaptable)Certifications Non-Mandatory: GCP Professional Cloud Developer / Microsoft Azure Certified Developer
Section V - Sample Questions for Training ModelTechnical:

1. Why do we use Fast API? How is it different from other frameworks?
2. What is a lambda function in Python, and how is it used?
3. What are the different data types available in Python?

Coding:

1. Implement a function to calculate the maximum sum of all subarrays of an array of integers.,
2. Write a Python function to retrieve elements of a list along with their corresponding indices
3. Explain the concepts of async and await in Python. How do they facilitate asynchronous programming?
4. What is the difference between concurrency and multi-threading in Python? Please explain the use case architecture for each.

berribot jd version 3

📌 Walk-in || Generative AI Developer (Hyderabad)
🏢 Tata Consultancy Services
📍 Hyderabad