Senior Custom Software Engineer - AI & Java Full Stack

Accenture India Private Limited; Bengaluru; On-site; INR 1,800,000 - 2,400,000; Full time

Job summary

Accenture India Private Limited is seeking a Custom Software Engineer for a project role in Bengaluru. This role focuses on designing and building AI-powered, multi-agent systems with LLMs, RAG pipelines, vector stores, and enterprise APIs.

You will lead architecture, framework choices, and governance while collaborating with cross-functional teams to deliver scalable, production-grade AI solutions. Strong CS fundamentals and 12+ years of experience are expected.

Qualifications

• Strong CS and software engineering fundamentals.
• Hands on experience with LLMs, RAG, vector DBs, and AI APIs.
• Ability to build end to end AI-powered applications and integrate AI into systems.

Responsibilities

• Define end to end agentic architectures (orchestration, planning, tool use).
• Publish reusable reference designs and patterns for multi agent workflows.
• Translate business problems into practical AI solutions with scoping and PoC.
• Own non functional requirements: security, privacy, performance, cost efficiency.
• Establish data contracts and retrieval patterns for RAG pipelines.

Skills

LLMs

RAG

Vector DBs

AI APIs

Agent orchestration

Education

15 years full time education

Tools

LangChain

LlamaIndex

Semantic Kernel

pgvector

Pinecone

Weaviate

Milvus

Kafka/JMS

API gateways

Job description

Custom Software Engineer Project Role

Project Role Description : Develop custom software solutions to design, code, and enhance components across systems or applications. Use modern frameworks and agile practices to deliver scalable, high-performing solutions tailored to specific business needs. Must have skills : Java Full Stack Development Good to have skills : NA Minimum 12 year(s) of experience is required Educational Qualification : 15 years full time education

Summary

Role Summary AI Powered Tech Talent Lead the design and delivery of agentic AI solutions—composable, multi agent systems that integrate LLMs, RAG pipelines, vector databases, and enterprise APIs—so teams can move from automation to autonomy across business workflows. AI Powered Tech Talent.

Outcomes & Responsibilities

• Architecture & Patterns
  • Define end to end agentic architectures (orchestration, planning, tool use, memory, evaluation loops), and publish reusable reference designs and patterns for multi agent workflows.
  • Standardize components: LLM providers, embedding models, vector DBs, retrieval strategies, prompt templates, evaluators, and observability stacks.
• Solution Design & Delivery
  • Translate business problems into practical AI solutions lead discovery, scoping, PoC, pilot, and scale out phases with clear communication to business and engineering stakeholders.
  • Own non functional requirements: security, privacy, compliance, performance, reliability, and cost efficiency (incl. token/compute budgeting).
• Data & RAG
  • Establish data contracts and retrieval patterns (chunking, indexing, hybrid search, citations/grounding, freshness).
  • Design guardrails to minimize hallucinations and enforce factuality.
• Platforms & Integration
  • Define integration strategies with enterprise systems, event streams, APIs/gateways, and CI/CD ensure alignment with SDLC and release governance.
• Quality & Responsible AI
  • Set evaluation frameworks (offline/online), E2E telemetry, failure modes/mitigations, and safety policies for content, access, and action boundaries.
• Leadership
  • Guide squads of developers contribute to internal enablement (playbooks, training tracks, coding standards).
  • Support interviewer panels for AI native hiring.

Must Have Qualifications (AI Native archetype)

Strong CS and software engineering fundamentals. Hands on experience with LLMs, RAG, vector DBs, and AI APIs. Proven ability to build end to end applications and integrate AI into systems. Visible portfolio of AI projects demonstrating experimentation and speed. Fast learner with a bias for rapid prototyping and iteration. Able to translate business problems into practical AI solutions with clear communication.

Programming Languages (Architect) Primary proficiency (at least one):

• Python (fast prototyping, LangChain/LlamaIndex, eval harnesses)
• TypeScript/Node.js (agent orchestration in services, API tooling)
• Java (enterprise integration, microservices, performance)

Secondary/beneficial:

• C#/.NET, Go, Rust (for performance critical agents or tooling)
• SQL (PostgreSQL/pgvector)
• Shell scripting (DevOps automation)

Ecosystem familiarity:

• LangChain, LlamaIndex, Semantic Kernel vector DBs (pgvector, Pinecone, Weaviate, Milvus) messaging (Kafka/JMS) API gateways cloud SDKs (Azure/AWS/GCP)

Preferred Skills & Experience

• Designing multi agent systems (planner/executor roles, collaboration protocols, tool abstraction) experience with agent frameworks or custom orchestration.
• Production grade RAG (retrieval evaluators, re ranking, freshness pipelines), vector DBs, and hybrid search.
• LLMOps/ModelOps: prompt/versioning, eval harnesses, feature flags, canarying, A/B testing, safety gateways.
• Cloud native architecture (containers, serverless, events), secrets management, policy as code, and cost governance for AI workloads.
• Enterprise integration: API gateways, event buses, messaging (JMS/Kafka), and contract testing.

Additional Information:

• - The candidate should have minimum 12 years of experience in Java Full Stack Development.
• - This position is based at our Bengaluru office.
• - A 15 years full time education is required.