Tanishq Chaturvedi profile photo 1 on Tal

Tanishq Chaturvedi

Generalist

Indore

1+ year of work experience

Dice
Aug 20251 yr 1 moPresent

Gurugram, Haryana, India

1. SAP-Integrated R2R Platform: Led the architecture and delivery of a Record-to-Report (R2R) automated accrual platform in close collaboration with PwC. Handled the distributed logic to ingest and compute metrics for 396,000+ PO lines and 960,000+ GRN transactions per fiscal year, preventing in-memory server degradation. This core engine successfully cut corporate processing costs for JKCements from ₹9 Crores down to a one-time ₹3 Crores, maintaining an ultra-lean recurring annual cost of just ₹57 Lakhs. 2. Production Voice AI (Tabbly Integration): Deployed an outbound multi-lingual conversational agent (Marathi, Hindi, English) with stateful checkpointing for the Mahavitaran Electricity Board. By engineering targeted prompt optimizations and context-minimization logic, we slashed round-trip voice latency by 19% and token overhead by 17%—reducing total runtime costs to a lean ₹1.59/min and yielding ₹14.36 Lakhs in annualized infrastructure savings across 3,000 daily calls. 3. Payabalegpt.ai In the FinTech layer of our product suite, PayabaleGPT focuses on the "Order-to-Pay" workflow. It leverages Generative AI to handle invoice processing, dispute resolution, and payment scheduling. The Impact: By removing the manual friction from accounts payable, we enable finance teams to focus on strategy rather than data entry. The Challenge: Ensuring 99.9% accuracy in financial data extraction and maintaining secure, compliant AI-driven financial pipelines.
Dice
Feb 20256 mosJul 2025

Pune District, Maharashtra, India

Engineer a chargeback ingestion microservice using Node.js, TypeScript, Express.js, PostgreSQL, BullMQ, and Redis to process large Excel/CSV file uploads, perform SHA-256 deduplication, and manage resumable jobs with Drizzle ORM. • Implement parallel and batched processing with map–reduce style workers, improving throughput by 240% and reducing job time by 71% for 100k-row files. • Build LangGraph multi-agent pipelines with fan-in/fan-out and checkpointing, increasing extraction throughput by 110%, re- ducing latency by 19%, and lowering token usage by 17%. • Add reliability and observability features including progress tracking, retries with dead letter queues, idempotent replays, and status APIs — achieving a 99.95% successful job completion rate for sustained 150k+ record loads, and reducing compute costs by 12%.
Lumber
Oct 20243 mosDec 2024

Bengaluru, Karnataka, India

• Developed the backend for an Equipment Management System supporting 30,000+ users across 60+ companies using Spring Boot, Java, JPA, and PostgreSQL. • Optimized PostgreSQL queries through indexing and advanced Spring Data JPA techniques, achieving 43% faster query performance for large datasets. • Built secure, scalable REST APIs with pagination and AWS S3-based file handling. • Delivered 7 full-stack features using TypeScript, React.js, and Redux, ensuring smooth frontend–backend integration.

Tanishq’s AI Stack

codex
codex
cursor
cursor
kimi
kimi
langgraph
langgraph
langsmith
langsmith
OpenRouter
OpenRouter

Education

Institute of Engineering & Technology DAVV, Indore

Bachelor of Technology - BTech, Computer Science

Institute of Engineering & Technology DAVV, Indore

20214 yrs2025
Agarwal Public School

PCM, School

Agarwal Public School

200514 yrs2019

Join Tanishq and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Tanishq Chaturvedi on Tal

Headline

Generalist

About

Hi, I'm Tanishq Chaturvedi! I am a Software Developer and Computer Science Engineering grad focused on building scalable backend architectures and stateful generative AI applications. What I Do: AI & Orchestration: Advanced Multi-Agent Meshes, LangGraph, LangChain, RAG, and production LLM optimization. Backend Systems: High-throughput distributed pipelines, microservices, and database optimization. Full-Stack Integration: Aligning robust TypeScript/React frontends with complex distributed backends. Tech Stack: Languages & DBs: TypeScript, JavaScript, C++, Java, SQL, PostgreSQL, Redis, Supabase, Pinecone Frameworks & Tools: Node.js, Express.js, Spring Boot, Drizzle ORM, Redux Toolkit, Docker, GCP

Location

Indore

AI stack

codex

cursor

kimi

langgraph

langsmith

OpenRouter

Work experience

Software Engineer at Dice, Aug 2025 - Present · 1 yr 1 mo, Gurugram, Haryana, India, 1. SAP-Integrated R2R Platform: Led the architecture and delivery of a Record-to-Report (R2R) automated accrual platform in close collaboration with PwC. Handled the distributed logic to ingest and compute metrics for 396,000+ PO lines and 960,000+ GRN transactions per fiscal year, preventing in-memory server degradation. This core engine successfully cut corporate processing costs for JKCements from ₹9 Crores down to a one-time ₹3 Crores, maintaining an ultra-lean recurring annual cost of just ₹57 Lakhs. 2. Production Voice AI (Tabbly Integration): Deployed an outbound multi-lingual conversational agent (Marathi, Hindi, English) with stateful checkpointing for the Mahavitaran Electricity Board. By engineering targeted prompt optimizations and context-minimization logic, we slashed round-trip voice latency by 19% and token overhead by 17%—reducing total runtime costs to a lean ₹1.59/min and yielding ₹14.36 Lakhs in annualized infrastructure savings across 3,000 daily calls. 3. Payabalegpt.ai In the FinTech layer of our product suite, PayabaleGPT focuses on the "Order-to-Pay" workflow. It leverages Generative AI to handle invoice processing, dispute resolution, and payment scheduling. The Impact: By removing the manual friction from accounts payable, we enable finance teams to focus on strategy rather than data entry. The Challenge: Ensuring 99.9% accuracy in financial data extraction and maintaining secure, compliant AI-driven financial pipelines.

Software Engineer Intern at Dice, Feb 2025 - Jul 2025 · 6 mos, Pune District, Maharashtra, India, Engineer a chargeback ingestion microservice using Node.js, TypeScript, Express.js, PostgreSQL, BullMQ, and Redis to process large Excel/CSV file uploads, perform SHA-256 deduplication, and manage resumable jobs with Drizzle ORM. • Implement parallel and batched processing with map–reduce style workers, improving throughput by 240% and reducing job time by 71% for 100k-row files. • Build LangGraph multi-agent pipelines with fan-in/fan-out and checkpointing, increasing extraction throughput by 110%, re- ducing latency by 19%, and lowering token usage by 17%. • Add reliability and observability features including progress tracking, retries with dead letter queues, idempotent replays, and status APIs — achieving a 99.95% successful job completion rate for sustained 150k+ record loads, and reducing compute costs by 12%.

Full Stack Developer at Lumber, Oct 2024 - Dec 2024 · 3 mos, Bengaluru, Karnataka, India, • Developed the backend for an Equipment Management System supporting 30,000+ users across 60+ companies using Spring Boot, Java, JPA, and PostgreSQL. • Optimized PostgreSQL queries through indexing and advanced Spring Data JPA techniques, achieving 43% faster query performance for large datasets. • Built secure, scalable REST APIs with pagination and AWS S3-based file handling. • Delivered 7 full-stack features using TypeScript, React.js, and Redux, ensuring smooth frontend–backend integration.

Education

Bachelor of Technology - BTech, Computer Science, Institute of Engineering & Technology DAVV, Indore

PCM, School, Agarwal Public School