RISHAV

AI |

I build production AI systems that reduce costs by 60% and turn raw data into business decisions.

8+Years
10+Brands
31Projects
Available for projects starting Q2 2026

Trusted by Industry Leaders

The Weather CompanyIBMBitwiseGartnerNovartisPolestarJubilant FoodWorksDominosReckitt BenckiserDettolIndiaMart

What I Do

AI System Architecture

End-to-end AI systems — from RAG pipelines to production LLM deployments.

Python · FastAPI · LangChain · RAG

Data Platform Consulting

Scalable data infrastructure that turns raw data into business decisions.

Spark · Airflow · dbt · Snowflake

AI Research & Evaluation

Benchmark, evaluate, and fine-tune models for your specific domain.

LLMs · Benchmarks · Fine-tuning

Selected Work

Case Study 01

Programmatic Ad Pipleline Optimization

Challenge

Legacy processing pipeline was taking 14+ hours, frequently missing internal SLAs and causing revenue attribution delays.

Solution

Re-architected the entire flow using AWS Batch for parallel processing and optimized Snowflake merge logic to reduce lock contention.

Impact

Reduced runtime to ~5 hours (60% improvement) and cut compute costs by 30% by moving off always-on instances.

60%Time Saved
30%Cost Reduction
99.9%SLA Adherence
Case Study 02

Enterprise Data Quality Framework

Challenge

Manual data validation across 15+ brands led to silent failures and bad data entering executive dashboards.

Solution

Built a serverless validation framework (AWS Lambda) with AI-generated rules that syncs bidirectional with Google Sheets for non-tech users.

Impact

Eliminated manual QA time by 100% and caught 50+ critical data issues in the first month before they reached production.

-100%Manual Effort
15 BrandsCoverage
50+Errors Caught
Case Study 03

GOAT - Gartner's Own Agentic Tech

Challenge

Gartner associates needed quick access to client information, engagement data, and research materials but relied on fragmented internal tools.

Solution

Built GOAT, a conversational agentic AI assistant with RAG pipelines, multi-agent orchestration, and re-architected infrastructure from AWS EKS to serverless Lambda.

Impact

Serving 5,000+ concurrent users with 88% query resolution improvement, 35% infrastructure cost reduction, and 40% token efficiency via A/B testing.

5,000+Users
+88%Resolution
35%Cost Saved

Rishav is an exceptional talent who consistently delivers high-quality solutions. His technical expertise and problem-solving skills make him an invaluable asset to any team.

Ivan Cheklin · BI Leader, IBM - TWC

Let's Build Something Together

Get In Touch

© 2026 RISHAV