Senior AI Engineer

Saurabh Chakrabarty Email : ***********@*****.***

Pune, India — Open to Relocation (Global) Mobile : +91-841******* LinkedIn, GitHub

Summary

Principal/Lead AI Engineer with 7+ years’ experience designing and deploying AI/ML systems, document intelligence, and knowledge graph pipelines with core focus on deterministic inference and workflow automation. Experienced across startups and enterprise systems, leading AI-native architectures, LLM orchestration, and automation-first workflows. Highly skilled in building scalable, privacy-first platforms by integrating leading frameworks, and deploying across AWS/Azure/GCP environments. Experience

•

Picard.law Remote

Co-founder & Lead AI Engineer April 2024 – Present

Built AI-first legal automation converting unstructured emails and contracts into structured knowledge graphs.

Architected graph + vector retrieval for compliance checks, document linkage, and lifecycle automation.

Scaled document pipelines for OCR, deduplication, and metadata extraction (1M+ pages).

Enabled local document processing + offline AI inference using SLMs for a couple of legal services companies.

•

Independent Consulting (Upwork & Direct) Remote

Principal AI Consultant April 2024 – Present

Delivered AI discovery and implementation consulting for Legal, Finance, and Document AI automation.

Designed agentic RAG and multi-agent systems using LlamaIndex, LangGraph, Mastra.ai, Agno.ai.

Led deployment for SoulBazaar.ai (now LetsAgent.ai): onboarding 1500+ users, deterministic social post generation from user context.

•

Orijin+ Hybrid — Perth, AUS

Senior Data/AI Engineer May 2021 – April 2024

Enabled large scale data warehouse for efficient analytics delivery on products sale and trends.

Built scalable IoT backend (Go, TypeScript) for tracking global shipments with ease deployed on AWS.

Enabled product recommendation system based on trends, and relevance to user’s history.

•

Botminds AI Chennai, India

AI Engineer Jan 2020 – Apr 2021

Trained segmentation models from scratch, with seq-to-seq accuracy over 85%.

Focused on large-scale document processing speed, while maintaining accuracy.

Built modular IDP workflows reducing manual review time by 70%.

Engineered baseline table structure extraction models, and pipelines, enabling new revenue streams.

•

SymphonyAI (Acquired Surukam Analytics) Chennai, India AI Engineer May 2018 – Dec 2019

Continued R&D post-acquisition of Surukam; worked on hand-detection & handwriting extraction models.

Integrated computer vision-based recognition into enterprise analytics stacks.

Worked on new crossover platform’s backend called Krystal AI, enabling heuristics and automation improving upon Surukam’s CruxIQ.

•

Surukam Analytics Chennai, India

Research Intern Nov 2017 – Apr 2018

Researched computer vision & hand-detection algorithms for scanned legal document processing.

Contributed to technology leading to company acquisition by SymphonyAI. Selected Projects & Achievements

• Email Knowledge Automation: Developed end-to-end pipeline converting enterprise email threads into searchable graph + vector knowledge bases.

• Agentic Compliance Checker: Built automated compliance validation pipeline using policy graph reasoning & RAG orchestration.

• Large-Scale Document AI: Architected high-volume document pipelines for OCR, deduplication, & signature extraction (1M+ pages).

• LetsAgent.ai (formerly SoulBazaar.ai): Implemented user-personalized post-generation pipeline using Mastra.ai and Twitter automation APIs.

• On-Prem Deployment (Proxxio / LegalEase): Implemented local document processing, knowledge graph based retrieval, and deterministic QnA with citations on legal documents (Unsloth + KV cache). Tools / Ecosystem

• Agentic Stack: LlamaIndex, LangGraph, Mastra.ai, Agno.ai, LangChain

• Cloud AI Platforms: AWS Bedrock, Google Vertex AI, Azure OpenAI

• Architecture-as-Code: Mermaid, Structurizr, PlantUML, diagrams.net, Lucidchart

• Core: Python, TypeScript, PyTorch, TensorFlow, scikit-learn, Neo4j, PostgreSQL, MongoDB, Redis, AWS, Azure, Vercel

Education

•

Fergusson College (University of Pune) Pune, India M.Sc., Industrial Mathematics & Computer Applications — GPA: 3.94 (9.86/10.0) July. 2015 – June. 2018

•

N. Wadia College (University of Pune) Pune, India

B.Sc., Mathematics — First class with distinction July. 2012 – June. 2015

Contact this candidate