Aadeesh Patni

Software Engineer

Bengaluru, Karnataka, India3 yrs 9 mos experience

Highly Stable

Key Highlights

Architected AI solutions impacting 200K+ users weekly.
Reduced onboarding time from 6 weeks to under 1 week.
Boosted document query accuracy significantly.

Stackforce AI infers this person is a SaaS-focused Software Engineer with expertise in AI and distributed systems.

Contact

Skills

Core Skills

Distributed SystemsJavaLarge Language Models (llm)React.js

Other Skills

TypeScriptAmazon Web Services (AWS)Python (Programming Language)Android DevelopmentData StructuresAlgorithmsKotlinC (Programming Language)Node.jsBack-End Web Development

About

SDE-2 at Amazon building AI products at scale — currently shipping enterprise GenAI features on Amazon Quick used by 200K+ weekly active users.

Experience

3 yrs 9 mos

Total Experience

3 yrs 9 mos

Average Tenure

3 yrs 9 mos

Current Experience

Amazon

2 roles

Software Development Engineer 2

Promoted

Jul 2024 – Present · 1 yr 10 mos

Architected a structure-aware document parsing pipeline for enterprise AI chat (Amazon Quick), preserving tables, images & complex layouts across PDF, Word, and PPT — boosting query accuracy by +12% (PDF), +20% (Word), +30% (PPT) across 2M+ uploads/week
Led end-to-end delivery of a Slack-native AI assistant with contextual chat, LLM-based Daily Digest notifications, and real-time streaming — scaling to 200K+ weekly active users
Designed and shipped real-time web search integration into Amazon Q Business via hybrid RAG, powering 34% of all chat queries and significantly reducing hallucinations on time-sensitive topics
Built a model-based evaluation framework to benchmark document parsing strategies, enabling data-driven pipeline selection with full security review and enterprise operational readiness

JavaDistributed Systems

Software Development Engineer

Jul 2022 – Jun 2024 · 1 yr 11 mos

Built core components of Amazon's internal GenAI HR assistant, including LLM-based intent routing using Claude 3.5 Haiku, RAG pipeline optimization via Contextual Retrieval, and a full WebSocket-based chat UI with rich React components
Redesigned the Change Management System architecture — remodeling configuration deployments as versioned, orchestrated workflows with approvals and rollback — reducing client onboarding from 6 weeks to under 1 week and migrating 7 major services