Project Role : Custom Software Engineer Project Role Description : Develop custom software solutions to design, code, and enhance components across systems or applications. Use up-to-date frameworks and agile practices to deliver scalable, high-performing solutions tailored to specific business needs.

Must have skills : Generative AI Good to have skills : AI Agents &

• Workflow Integration Minimum 5 Year(s) Of Experience Is Required Educational Qualification : 15 years full time education Core Engineering &

• Day to Day Practice Strong Python fundamentals Hands-on experience using Claude Code for day-to-day development and coding workflows Agentic Systems &

• Architecture Proficiency with agentic frameworks (e.g., LangGraph, custom orchestration patterns) — not just chatbot/RAG experience Experience selecting, integrating, and evaluating LLM APIs (e.g., Anthropic, OpenAI, others), including assessing tradeoffs across models and providers Proficiency with agentic planning and tool-use patterns (ReAct, Chain-of-Thought + action) experience building and shipping multi-agent pipelines in production, including handling failure modes, loops, and hallucinations Ability to design prompts systematically as part of end-to-end agent behavior Comfort with async, event-driven architectures Memory &





• Retrieval Architecture Experience designing agent memory architectures, including short- and long-term memory, embeddings, vector databases, and context window/token management Experience building and operating production RAG systems, including horizontally scalable vector databases, with attention to latency, reliability, and throughput tradeoffs Evaluation &

• Human in the Loop Ability to define task-specific evals and regression tests to catch prompt drift, retrieval failures, and behavioral regressions Understanding of online vs. offline evaluation tradeoffs and using results to drive iterative improvements Experience with human-in-the-loop (HITL) design and evaluation, including approval gating, routing ambiguous outputs to humans, and closing the feedback loop (e.g., Langfuse human annotation evals) Integration, Governance &

• Security Experience with MCP server integration Experience integrating agents with enterprise systems (internal APIs, identity/auth layers) Understanding of AI governance: audit trails, explainability, access controls, and data residency constraints Security-conscious agent design: scoped permissions and prompt injection mitigation Bonus: Experience operating within IT governance or change management processes

📌 Custom Software Engineer (Hyderabad)
🏢 Accenture in India
📍 Hyderabad