Principal Engineer

About Us

Position Name

Python Full Stack Engineer – Argus Integration (GenAI/Agentic AI)

Experience

5–8 years of professional software engineering experience, with strong hands-on expertise in Python and practical experience building LLM/GenAI applications.

Location

Bangalore

Shift Timings

General

Job Description

•    Design and build agent tools and orchestration on AWS Strands Agents and Bedrock — both deterministic rule tools and LLM-judgment tools grounded in retrieval.
•    Implement RAG pipelines end to end: ingestion, chunking, embedding, retrieval, and grounding — including knowledge-graph-anchored retrieval where it improves precision.
•    Write evaluation and test harnesses — deterministic checks, LLM-as-judge evaluators, and regression suites — to measure faithfulness, correctness, and safety.
•    Add guardrails and human-in-the-loop review hooks appropriate for a compliance-sensitive product.
•    Build clean integrations with platform and third-party systems (e.g. document sources, Veeva Vault, downstream services).
•    Instrument agent behaviour into the observability stack (Langfuse, OpenTelemetry/ADOT) and use traces to debug and optimize cost and latency.
•    Participate actively in code review, design discussions, and documentation; help level up peers and improve team practices.
•    Ship production-quality code with strong testing, error handling, and reproducibility as defaults, not afterthoughts.

Required qualifications

•    5–8 years of professional software engineering with strong, current Python.
•    Hands-on experience building LLM/GenAI features — RAG, prompt engineering, and agent/tool-use (function calling).
•    Familiarity with at least one agent framework (Strands Agents, LangGraph, LlamaIndex, AutoGen, CrewAI, or equivalent) and one LLM platform (Bedrock, OpenAI, Anthropic, Vertex).
•    Solid software fundamentals: data structures, API design, testing, version control, and CI/CD.
•    Comfort deploying and running services on AWS (or a comparable cloud).
•    A quality mindset — you test, measure, and reason about LLM correctness rather than eyeballing outputs.
Preferred / nice-to-have
•    Vector databases (pgvector, Pinecone, Weaviate, Qdrant, Chroma) and retrieval tuning.
•    Neo4j / knowledge graphs and GraphRAG.
•    Evaluation frameworks (RAGAS, DeepEval, promptfoo, or the Strands Evals SDK) and LLM-as-judge patterns.
•    Observability for LLM apps (Langfuse, LangSmith, OpenTelemetry/ADOT).
•    Exposure to a regulated or human-in-the-loop domain (pharma, healthcare, finance).

Competencies

Key Skills

Python, Generative AI, LLMs, RAG, Prompt Engineering, Agentic AI, AWS Bedrock, Strands Agents, API Integration, Argus Integration, Veeva Vault, Vector Databases, Knowledge Graphs, GraphRAG, Langfuse, OpenTelemetry, Automated Testing, CI/CD.

Soft Skills

• Strong analytical and problem-solving abilities.
• Excellent verbal and written communication.
• Collaboration and teamwork.
• Attention to detail and commitment to quality.
• Adaptability and willingness to learn emerging technologies.
• Ownership, accountability, and a proactive approach to resolving issues.
• Ability to communicate technical concepts clearly to business and technical stakeholders.

Qualification

Bachelor’s or Master’s degree in Computer Science, Engineering, Information Technology, or a related field, with 5–8 years of relevant professional software engineering experience.

Certifications