Accenture · Senior AnalystAnantapur, India

Sree Sai Bhavesh Veerabomma

Senior AI Engineer — Agentic Systems, RAG & LLM Platforms

I design and ship production-grade AI systems—retrieval, agents, evaluation, and guardrails—measured end to end.

40%

less manual work

25%

better retrieval

30%

lower token costs

70%

fewer unsafe outputs

Experience

2022 — present

AccentureJun 2026 — Present

Architecting agentic AI platforms for enterprise banking, combining multi-agent reasoning, deterministic policy controls, and human approvals.

TemporalLangGraphMCPNeo4jGoRulesFastAPI

AccentureJun 2024 — May 2026

Built production RAG systems, evaluation pipelines, safety guardrails, APIs, and observability for high-stakes financial workflows.

PythonRAGRAGASDeepEvalNemo GuardrailsPrometheus

HawkDigital · InternshipJan 2024 — Feb 2024

Deployed a 24/7 AWS scraping pipeline for 80,000+ real-estate records and reporting that helped secure five client accounts.

PythonSeleniumAWSData pipelinesDashboards

Omdena · InternshipJun 2022 — Sep 2022

Collected, cleaned, and validated structured data for 5,000+ secondary schools from multiple public sources.

PythonSeleniumBeautifulSoupData qualityAnalysis

Production financial AI workflow

Problem: Credit appraisal at an enterprise bank was slow, manual, and hard to audit end to end.
Built: A multi-agent appraisal workflow combining reasoning agents, deterministic policy controls, and human-in-the-loop approvals.
Impact: Cut manual effort by 40% and unsafe model outputs by 70%.

LangGraphGoRulesNeo4jTemporalFastAPI

Retrieval, evaluation, guardrails, observability

Problem: Financial workflows needed grounded answers with provable, measurable quality.
Built: Production RAG pipelines with an evaluation harness, safety guardrails, and full observability.
Impact: 25% better retrieval and 30% lower token costs.

PythonRAGRAGASDeepEvalNemo GuardrailsPrometheus

LangGraph + Google Workspace agent ecosystem

Problem: Specialist tasks were scattered across Google Workspace tools with no shared brain.
Built: A conversational LangGraph platform routing autonomously across six specialized agents with 25+ Workspace integrations, persistent memory, routing validation, loop prevention, and audit logs.
Impact: 30% more reliable multi-agent workflows.

LangGraphFastAPIMCPOracle DBAsync Python