Tanishk Shah

Product Manager

Mumbai, Maharashtra, India3 yrs 8 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

5X Salesforce Certified professional
Expertise in Large Language Models and AI
Proven experience in developing automated systems

Stackforce AI infers this person is a skilled AI and Machine Learning developer with a focus on healthcare and fintech solutions.

Contact

Skills

Core Skills

Large Language Models (llm)Artificial Intelligence (ai)Machine Learning

Other Skills

Python (Programming Language)FastAPIREST APIsDatabasesModel TrainingAgentsOptical Character Recognition (OCR)Deep LearningJavaSQLProgramming Languages

Experience

3 yrs 8 mos

Total Experience

1 yr 10 mos

Average Tenure

2 yrs 6 mos

Current Experience

Deloitte

DC Analyst

Oct 2023 – Present · 2 yrs 6 mos · Mumbai, Maharashtra, India · On-site

GenAI Engineer | LLM Observability | Guardrails | Evaluation/Metrics | AWS | Python/FastAPI
GenAI Engineer/Developer delivering enterprise GenAI platform capabilities for Pfizer (current) and previously Eli Lilly & Co. I build production LLM systems that are observable, governed, safe, and scalable.
LLM & AI
Transformers (GPT/BERT), RAG, Knowledge Graphs
Vector search (FAISS/Pinecone) and semantic evaluation
LLM Observability (Langfuse) + LLM Gateway
Integrated Langfuse into LLM gateway/router flows (LiteLLM) to capture traces/spans/generations
Automated onboarding: org + default projects, validations, and bulk org creation scripts
Built dashboards and observability documentation (architecture + tracing model)
Improved reliability: safe execution when observability is off, Redis hardening with DynamoDB fallback/sync, multi-region readiness
Safety & Guardrails
Implemented NVIDIA NeMo Guardrails for safety/compliance
Added guardrail signals to traces; disabled tracing when guardrails aren’t configured
Built guardrails catalog APIs
Evaluation, Metrics & Governance
Built real-time metrics capture and bulk evaluation pipelines (async/Celery workers)
Metrics Library (add/edit/activate) with Alembic/PostgreSQL + APIs
Governance logging: LLM transactions, metering, audit logs
Run governance (real_time vs bulk_evaluation), strong validations; supported UAT/prod fixes
Alerts + notifications: thresholds, scheduled jobs (EventBridge/Lambda patterns), bulk-eval completion emails
Data retention (incl. usecase-level config) with auth/validation
Custom LLM provider option for metrics execution (OpenAI-format base URL + API key)
SonarQube/linting fixes, shared-utils refactors, deployment support
✅ Pfizer – Observability, guardrails, evaluation/metrics, logging, alerts, retention
✅ Eli Lilly – LLM-driven pharma data analysis & clinical research automation

Large Language Models (LLM)Python (Programming Language)FastAPIArtificial Intelligence (AI)Machine LearningREST APIs+3

Sbi life insurance co. ltd.

Software Developer

Jan 2023 – May 2023 · 4 mos · Mumbai, Maharashtra, India · On-site

Worked as a Software Developer Intern, gaining hands-on experience in image processing, OCR-based data extraction, and AI-driven document verification.
Developed an automated KYC verification system, extracting user information from uploaded identity documents during the onboarding process on the SBI Life Insurance portal.
Implemented Optical Character Recognition (OCR) techniques using Tesseract OCR, OpenCV, and deep learning models to accurately extract and validate personal details from scanned images.
Designed and trained a Machine Learning model to verify the authenticity of identity documents (PAN and Aadhaar cards) by detecting anomalies, ensuring compliance with regulatory standards.
Leveraged image generation and enhancement techniques to compare submitted documents against reference templates, improving fraud detection and data validation.
Applied convolutional neural networks (CNNs) and computer vision algorithms for document classification and forgery detection.
Integrated Python-based automation scripts for data preprocessing, feature extraction, and real-time document verification.