Open to opportunities

Pratik Abnave

@pratikabnave

Message

AI Engineer building production GenAI, RAG, and scalable data pipelines for enterprise insights.

India

Message

What I'm looking for

I’m looking for a team where I can build production-grade GenAI and ML platforms—RAG, agent workflows, and robust MLOps—with real impact in finance/enterprise, strong engineering culture, and room to scale intelligent products.

I’m an AI Engineer with 4+ years of experience building data-intensive research platforms, AI applications, and scalable data pipelines across financial and enterprise domains. I focus on turning complex datasets and documents into actionable insights using production-grade AI, analytics, and workflow-driven delivery.

In my current role, I architected an AI CRM Intelligence Platform that replaced analyst-dependent reporting with self-service natural language queries, deployed across 23 MongoDB collections via 5 coordinated AI agents and hybrid retrieval (BM25 + dense embeddings). I also built 7 production machine learning models inside a 10-stage Apache Airflow DAG to automate RFM segmentation, revenue forecasting, and recommendations on every ingestion cycle, and engineered an autonomous voice AI agent that runs live CRM demos using browser automation and real-time RAG with crash recovery in under 3 seconds.

I’ve strengthened reliability and trust through AI governance and LLMOps guardrails, including destructive-keyword blocklists and read-only tenant isolation policies. Previously, as a Data Scientist, I developed transformer-based financial intelligence pipelines, built peer-comparison/benchmarking for 2,000+ companies (reducing turnaround by 30%), and improved forecasting accuracy by 40% with monitoring and retraining workflows; I also reduced data processing time by 40% and improved data reliability by 35% while enabling data-driven decisions at scale.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

Doodleblue Innovations

Jul 2025 - Present (11 months)

Architected an AI CRM Intelligence Platform enabling self-service natural language queries across 23 MongoDB collections using hybrid RAG and 5 coordinated AI agents, deployed company-wide and used weekly. Built 7 production ML models in a 10-stage Apache Airflow pipeline and implemented LLM governance/LLMOps guardrails enforcing read-only tenant isolation for GPT-4o queries.

Python Apache Airflow MongoDB Pinecone RAG AI Agents LLMOps Model Governance AI Observability

Data Scientist

Sylvr

May 2023 - Jul 2025 (2 years 2 months)

Developed NLP and transformer-based financial intelligence pipelines using Python, SQL, and FinBERT to extract metrics from analyst reports and regulatory filings. Built benchmarking and forecasting solutions for 2,000+ public companies and improved forecast accuracy by 40% using ARIMA, XGBoost, and LSTM, plus drift detection/monitoring that reduced performance degradation incidents by 40%.

Python SQL FinBERT Transformers ARIMA XGBoost Azure ML Model Monitoring

Data Scientist

Turing.com

Jun 2022 - Apr 2023 (10 months)

Mined, validated, and transformed 10M+ records across heterogeneous sources using Python, SQL, and Mode Analytics, improving data reliability by 35% and reducing processing time by 40%. Developed reusable ingestion/cleansing and feature-engineering frameworks (schema validation and monitoring) and deployed ML models on AWS/Azure with drift detection, improving predictive accuracy by 15%.

Python SQL Mode Analytics Data Validation Feature Engineering AWS Azure Monitoring Drift CI CD

Education

Degrees, certifications, and relevant coursework

University of Mumbai

Master of Computer Applications (MCA), Computer Applications

Completed a Master of Computer Applications (MCA) degree under the University of Mumbai through Late Bhausaheb Hiray Institute of Computer Application.