top of page
Scienaptic_RGB_Color.png
Full Tine

Software Engineer

Bengaluru, Karnataka, India

The Software Engineer will develop single- and multi-agent workflows using reasoning loops (e.g., ReAct, Chain-of-Thought). Build, test, and optimize advanced Retrieval-Augmented Generation (RAG) pipelines, embedding strategies, and prompt chains.

Responsibilities

  • GenAI & Agent Development (80%): Develop single- and multi-agent workflows using reasoning loops (e.g., ReAct, Chain-of-Thought). Build, test, and optimize advanced Retrieval-Augmented Generation (RAG) pipelines, embedding strategies, and prompt chains.

  • Data Engineering & Pipeline Support (20%): Build and maintain lightweight, reliable ETL/ELT pipelines to ingest, clean, and structure enterprise data for vector databases, knowledge graphs, and LLM consumption.

  • API & Tool Integration: Expose agent capabilities via robust REST APIs and connect models securely to internal database systems, enterprise software, and third-party services.

  • AWS Cloud Infrastructure: Deploy, manage, and scale AI workloads and data applications within the AWS ecosystem (e.g., ECS, Lambda, Bedrock, S3, RDS).

  • Evaluation & Reliability: Implement monitoring and tracing tools to log agent steps, track latency, manage API token spend, and mitigate hallucinations in production.

Skills & Competencies

  • Experience: 1–2 years of hands-on technical experience blending Generative AI development with core data infrastructure.

  • Programming: Strong proficiency in Python and fundamental software engineering best practices (Git, unit testing, clean code).

  • GenAI Stack: Experience with frameworks like LangChain, LlamaIndex, LangGraph, or CrewAI, alongside foundational knowledge of vector databases (e.g., Pinecone, Milvus, pgvector).

  • Data & APIs: Practical experience building and consuming RESTful APIs (using FastAPI, Flask, etc.) and writing structured SQL queries to manipulate relational databases.

  • AWS Cloud: Foundational experience working within AWS to deploy containerized applications or serverless functions.

Scienaptic is the world's leading AI powered Credit Underwriting platform company. Designed by seasoned Chief Risk Officers, its platform is creating industry leading business impact in terms of lifts such as higher approvals (15-40%) and lower credit losses (10-25%) with all the regulatory explain ability. Last year alone, we have helped financial institutions evaluate 45 Million consumers and offer credit to over 15 Million. Scienaptic clients include Fortune 100 banks, community banks and FinTech’s

bottom of page