Sagar Lakhani
Lead AI Engineer | LLM, RAG, ML & AI Agents | Building Scalable AI Products with Python | Algo Trader
- Role
- Development Team Lead at Vinfotech
- Location
- Indore, MP, IN
- LinkedIn followers
- 500 followers
About Sagar Lakhani
I am a Lead AI Engineer and Backend Architect with 7+ years of professional experience building production-grade AI systems, LLM-powered applications, and data-driven platforms across fintech, SaaS, and sports analytics.My core expertise lies in designing Retrieval-Augmented Generation (RAG) systems, AI agents, and LLM orchestration layers that enable users to interact with complex databases, documents, and real-time data through natural language. I specialize in building high-performance AI backends using Python, FastAPI, LiteLLM, and vector databases (pgvector, FAISS, Pinecone) to deliver low-latency, cost-optimized, and scalable AI solutions.I have led the development of:LLM-powered sales and support agents deployed across web, WhatsApp, and TelegramAI chatbots for fantasy sports and financial analytics capable of querying large SQL and NoSQL datasetsPredictive ML systems using TensorFlow, ScikitLearn, and Keras for market trends and player performance forecasting.I work extensively with PostgreSQL, MySQL, MongoDB, Redis, RESTful APIs, OAuth/SSO, Docker, and CI/CD pipelines, delivering enterprise-grade AI platforms that scale to active users.My background in Embedded Systems (M.Tech – DAVV | PG Diploma – CDAC ACTS Pune) provides a strong foundation in systems engineering, performance optimization, and hardware-software integration, which I admit to building reliable, production-ready AI infrastructure.I am passionate about transforming LLMs into real-world products, not demos — combining AI, backend engineering, and data systems to build high-impact, commercially viable platforms.
Experience
Development Team Lead
Jun 2025 — Present · Indore, IN
Built LLM-powered Sales & Support Agents using RAG pipelines, vector databases (Qdrant, FAISS), and LiteLLM for multi-model routing, cost optimization, and high-availability inferenceDesigned custom LLM orchestration layers in Python + FastAPI, enabling dynamic model selection, tool calling, memory, and multi-turn reasoning without framework lock-inDeveloped AI chatbots and analytics platforms for fantasy sports, fintech, and SaaS, enabling users to query complex SQL and NoSQL databases using natural languageImplemented hybrid RAG architectures combining structured (SQL/NoSql) and unstructured (documents, APIs, scraped data) sources with semantic searchIntegrated Google OAuth / SSO, role-based access, and secure session management for enterprise deploymentsDelivered predictive ML models (Keras, ScikitLearn) achieving 85%+ accuracy for market trends and player performance forecasting
Education
Rajeswar Vidhyalaya, Mhow
High School
1999 — 2009
Ujjain Engineering College, Sanwer Road, Ujjain
Bachelor’s Degree, Electrical, Electronics and Communications Engineering
2009 — 2013
CDAC ACTS PUNE
PG DIPLOMA, Embedded Systems and Design
2014 — 2015
School of Electronics, DAVV, Indore
Master’s Degree, Embedded Systems
2015 — 2017
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.