Akash Bakshi – AI Engineer | Public Profile on Dr.Job

About Me

AI Engineer specializing in Agentic AI, RAG systems, LLMs, and cloud-native deployments. Proven track record in building enterprise-scale intelligent systems — including an Agentic AI conversational platform for Canadian…

Experience

AI Engineer

MMNovaTech · Noida , Uttar Pradesh , India

Jul 2025 - Present · 1 year

Designed and developed an enterprise-grade Agentic AI conversational platform using LangGraph, enabling real-time, multi-turn interactions for nationwide property search, lead engagement, and customer assistance across Canada. Architected persistent conversational memory using MySQL, managing chat history, session states, user intents, and tool execution logs to support analytics, auditing, and CRM synchronization. Built tool-augmented AI agents for email automation, appointment scheduling, notifications, property search, and web search integrated with an enterprise CRM to streamline lead management and boost agent productivity. Built asynchronous AI workflows using FastAPI and Python to handle concurrent user requests with low latency.

AI Developer

Solitaire Infosystem · Chandigarh , Chandigarh , India

Jan 2025 - Jun 2025 · 5 months

Deployed an opensource LLM for high-quality image and text generation using an on-premises inference pipeline, reducing inference costs. Built a FastAPI backend for end-to-end request handling, real-time external API integration, LLM-powered content generation, and prompt-based template editing. Optimized the inference architecture to reduce latency and improve the speed of image, html, and text generation. Engineered a dynamic template rendering system for production-ready content delivery using prompt engineering.

AI Developer

Jspider Pvt Ltd · Chandigarh , Chandigarh , India

Feb 2024 - Dec 2024 · 9 months

Developed a web scraping pipeline using Python and BeautifulSoup to extract and preprocess data from multiple websites for building a searchable knowledge base. Designed a document ingestion pipeline with content cleaning, intelligent chunking, and semantic embeddings, storing vectors in Pinecone for efficient similarity search. Optimized AI inference pipeline for low-latency image, HTML, and text generation across multiple services. Engineered prompt-driven template generation and rendering pipeline for dynamic, production-ready content delivery.

Skills

Python Node.js Generative AI Agentic AI RAG Models Machine Learning Deep Learning NLP LLMs LSTM XGBoost Scikit-Learn OpenAI APIs Open Source LLM AWS FastAPI Fastify PineCone LangChain LangGraph PostgreSQL MySQL Docker Git Prompt Engineering Predictive Forecasting Power BI Data Analysis Web Scraping Excel BeautifulSoup GPT-4o API Twilio Asynchronous Input/Output HTML Template Generation API