AI Engineer with 2+ years in Generative AI, RAG, & Agentic Systems
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI/ML Engineer skilled in building end-to-end intelligent applications, from Agentic AI and RAG pipelines using Python, LangChain, and PyTorch to responsive frontends with React.js. Achieved 15x efficiency gains by integrating LLMs into seamless user experiences on Azure and AWS. Focused on creating scalable, user-centric solutions that merge AI innovation with robust software engineering principles.
Anantrao Pawar College of Engineering and Research
Bachelor of Computer Engineering · Computer Engineering
August 1, 2020 – May 1, 2024
Right & Left Brain Technologies
Senior Software Engineer
September 1, 2025 – Present
Pune, Maharashtra, India
Right & Left Brain Technologies
Software Engineer
July 1, 2024 – August 1, 2025
Pune, Maharashtra, India
compliance automation platform for Complifyre
September 1, 2025 – Present
Complifyre.ai is an AI-powered compliance automation platform Architected an Advanced RAG (Retrieval Augmented Generation) System using LangChain and Azure OpenAI (GPT-40) to automate compliance audits, reducing manual workload by 70% and increasing audit throughput by 3x. Engineered a high-precision Data Ingestion Pipeline combining OCR (Azure Document Intelligence) with Recursive Character Chunking and Embeddings (text-embedding-3-small) to extract granular evidence from complex PDF guidelines with 85%+ retrieval accuracy. Implemented Hybrid Search (Keyword + Semantic) using PostgreSQL (pgvector) and Cross-Encoder Re-ranking to minimize hallucinations, ensuring the model references the correct regulatory clauses with under 1 second retrieval latency. Deployed a scalable Architecture on Azure Cloud using Docker, enforcing RBAC for data security and implementing Redis Caching for frequently accessed compliance rules, achieving 99.9% uptime.
Agentic AI System for Kanalytics
September 1, 2025 – Present
An Agentic AI system developed for media monitoring Engineered a Multi-Agent Orchestration Framework using GPT-40 and ReAct (Reason + Act) prompting strategies, enabling autonomous agents to plan, scrape, and analyze media streams, achieving a 15x increase in processing throughput. Deployed specialized Task-Specific Agents for Named Entity Recognition (NER), Sentiment Analysis, and Topic Extraction, utilizing Few-Shot Prompting and Chain-of-Thought reasoning to surpass legacy systems with >95% classification accuracy. Built an Asynchronous Event-Driven Pipeline (using Redis/Celery) to handle 5,000+ articles daily, implementing Semantic Caching (caching embeddings vs. raw text) to reduce redundant LLM API calls by 40% and latency by 60%. Led the End-to-End MLOps Lifecycle, integrating Guardrails for output validation and Fine-tuning (LoRA) on smaller open-source models (e.g., Llama 3) for specific tasks, lowering cloud infrastructure costs by 25% while maintaining enterprise-grade security.
View ProjectMulti-Agent-AI-System
September 1, 2025 – Present
Autonomous Multi-Agent Orchestration System (Medical & Research) Engineered a custom Multi-Agent Orchestration Framework from scratch (without LangChain/CrewAI), implementing a Manager-Worker architecture to coordinate specialized agents for medical summarization and research drafting, demonstrating deep understanding of LLM state management and control flow. Implemented Agentic Reflection & Self-Correction loops by pairing primary agents with Validator Agents (e.g., PHI Sanitizer + Validator), reducing hallucinations and ensuring 100% compliance with sensitive medical data handling requirements. Designed complex Prompt Chaining workflows where output from one agent (Research Writer) serves as context for another (Refiner Agent), optimizing Context Window usage and improving the coherence of long-form generated content. Integrated robust observability using Loguru to trace Agent Reasoning steps (Chain-of-Thought) and API latency in real-time via a Streamlit dashboard, facilitating rapid debugging of agent behaviors.
View ProjectIntroduction to Large Language Models
June 1, 2026 – Present
Virtual Experience Program Participant
Skyscanner
June 1, 2026 – Present
Google Cloud Fundamentals: Core Infrastructure
June 1, 2026 – Present
Introduction to Generative AI
June 1, 2026 – Present
Cultural Fit Analysis
The candidate demonstrates a strong cultural fit for an AI Engineer role through their diverse project portfolio, which includes both professional and personal projects. The personal project, 'Multi-Agent-AI-System,' showcases initiative, deep technical curiosity, and a drive to build complex systems from scratch, which are highly valued in innovative tech environments. Their experience spans various aspects of AI development, from core model integration to scalable deployment and MLOps, indicating adaptability and a broad skill set. The focus on efficiency gains and cost reduction aligns with business-oriented development practices.
Soft Skills & Operational Fit
The candidate's project descriptions highlight strong problem-solving skills, particularly in optimizing complex AI systems for performance and cost. Their experience in leading end-to-end MLOps lifecycles and collaborating with cross-functional teams indicates good operational fit and teamwork capabilities. The detailed descriptions of architecting and engineering solutions suggest a structured and analytical approach to problem-solving.