Experience Requirements
This role requires a minimum of 4 years of hands-on AI/ML development experience (typically 4–5 years). Experience in Generative AI, LLM systems, and agentic AI development is fully counted toward this requirement. Relevant experience may include any combination of:
· Machine learning and deep learning engineering (model development, training, deployment).
· NLP and applied AI application development.
· Generative AI and LLM application development (RAG, prompt engineering, fine-tuning, embeddings).
· Agentic AI system development (tool use, planning, multi-agent orchestration, agent frameworks).
Note: Pure data analytics, BI/reporting, or infrastructure/support roles without hands-on AI/ML development do not count toward this requirement.
Key Responsibilities
Agentic AI Solutions
· Design, build, and deploy AI/ML pipelines and production-grade agentic workflows.
· Architect multi-agent systems employing orchestration patterns such as router, supervisor, hierarchical, and sequential pipeline models.
· Implement the agentic execution loop — reasoning, planning (ReAct), tool/function calling, memory management, and self-correction.
· Develop agents with human-in-the-loop checkpoints, guardrails, permissioned tool access, and auditable action trails.
· Integrate agents with enterprise systems, including relational databases, REST APIs, e-mail, ticketing, and notification platforms.
Knowledge Graph Engineering
· Design and build knowledge graphs: schema/ontology definition, entity and relationship extraction (LLM-based triple extraction, NER), entity resolution, and graph loading.
· Implement graph pipelines on graph databases such as Neo4j (Cypher), including modelling, ingestion, and query optimization.
· Expose knowledge graphs to AI agents as queryable tools, with schema descriptions and read-only, guardrailed access.
· Build GraphRAG capabilities — combining graph traversal with vector retrieval to support multi-hop reasoning and relationship-aware answers.
Full-Stack Engineering
· Develop full-stack applications with Python backends (FastAPI, Flask/Django) and modern frontends (React, Next.js, or similar frameworks).
· Build RAG systems, LLM integrations, and multi-agent orchestration layers.
· Write and optimize complex queries on enterprise databases.
· Write clean, testable, and well-documented code; participate in code reviews and own production reliability.
· Troubleshoot issues, optimize performance, and ship reliably.
Data Handling & Data Engineering
· Design and operate data pipelines (batch and streaming) that feed AI systems — covering ingestion (Airflow/Dagster), transformation (dbt/Spark), and storage (S3, data warehouses).
· Work across SQL, NoSQL, and graph stores — PostgreSQL, MySQL, MongoDB, and Neo4j — including schema design, query optimization, and data modelling for AI workloads.
· Implement change-data-capture (CDC), data validation (e.g., Great Expectations), and data-quality controls to ensure trustworthy AI outputs.
· Engineer pipelines for unstructured data, including document/PDF parsing, chunking, and metadata enrichment.
Data Readiness for AI & Agents
· Build end-to-end RAG pipelines: chunking strategies, embedding generation, vector storage (pgvector, Pinecone, Weaviate, ChromaDB), hybrid retrieval, and reranking.
· Prepare structured data for agent consumption — semantic layers, schema documentation, text-to-SQL validation, and entity-relationship/join modelling for safe and accurate agent querying.
· Design context-assembly mechanisms for LLMs — prompt construction, context-window management, and long-term agent memory (short-term state plus vector-based and graph-based recall).
· Define data freshness, re-indexing, and evaluation frameworks covering retrieval accuracy and answer faithfulness (e.g., RAGAS, TruLens).
Cloud, DevOps & Deployment
· Deploy and manage workloads on AWS, Azure, or GCP using containers, serverless architectures, and Infrastructure as Code (IaC).
· Manage containerization and orchestration using Docker and Kubernetes.
· Operate CI/CD pipelines, Git workflows, and DevOps fundamentals in a production setting.
Must-Have Qualifications
· Minimum 4 years (typically 4–5 years) of hands-on AI/ML development experience, inclusive of Generative AI, LLM, and agentic AI work.
· Strong Python proficiency, including FastAPI, Flask/Django, LangChain, LlamaIndex, Hugging Face, and PyTorch/TensorFlow.
· Hands-on experience building production agentic AI systems — tool use, planning, memory, and orchestration (LangGraph/CrewAI).
· Knowledge-graph building experience — schema/ontology design, entity–relationship extraction, entity resolution, and implementation on graph databases (Neo4j/Cypher).
· Full-stack development experience, specifically with APIs and frontends such as React, Next.js, or similar frameworks.
· Enterprise RDBMS expertise (PostgreSQL, MySQL, SQL Server, or Oracle), including schema design, query optimization, stored procedures, and data modelling.
· Familiarity with vector databases such as Pinecone, Weaviate, ChromaDB, or pgvector.
· Working knowledge of data-engineering fundamentals: ETL/ELT pipelines, data quality, and preparation of structured and unstructured data for LLM consumption.
· Experience with cloud platforms, preferably AWS (EC2, Lambda, S3, SageMaker); Azure or GCP are acceptable.
· Knowledge of Docker and Kubernetes for containerization and orchestration.
· Solid grasp of CI/CD pipelines, Git, and DevOps fundamentals.
· Expertise in LLM fine-tuning, embeddings, prompt engineering, and agentic AI concepts — function calling, ReAct, agent state, guardrails, and observability/tracing of agent runs.
Good-to-Have Qualifications
· Experience with multi-agent frameworks such as CrewAI, AutoGen, or custom orchestration.
· Agentic RAG experience — agents that self-query, rewrite queries, and iteratively refine retrieval.
· GraphRAG implementations — combining graph traversal with vector retrieval for relationship-aware, multi-hop question answering.
· Familiarity with document databases (MongoDB); RDF/OWL/SPARQL-based ontology engineering is a plus.
· Streaming data pipelines (Kafka/Kinesis) feeding real-time AI agents.
· Previous SaaS or multi-tenant platform experience.
· Knowledge of API gateway design, OAuth2/JWT, and microservices architecture.
· Prior experience working with government or enterprise clients.
What We Value
· A focus on shipping working software rather than creating slide decks.
· An ownership mindset where you see a problem and fix it.
· A commitment to staying current with AI tooling and trends.
· The ability to communicate clearly with both technical and non-technical stakeholders.
Details
· Location: Technopark, Trivandrum (on-site preferred; hybrid available for exceptional candidates).
· Type: Full-time.
· Send your updated resume to careers@gaudesolutions.com
