Headline

AI Engineer @ Sears | LLMs • Voice AI • RAG • Agentic AI • Generative AI • Python | Ex-Data Scientist @ Fidelity| IIT KGP’25

About

AI Engineer building production LLM and Voice AI systems. I enjoy solving the engineering problems that begin after the demo works—latency, evaluation, reliability, and deployment at scale. At Sears, I contributed to a voice AI scheduling platform that handles 2M+ calls and independently own a recruiting platform that runs 60+ voice interviews a day. I care about the parts most people skip: keeping p95 latency under 1.2s, building evaluation frameworks that actually catch a bad agent response, and knowing when cascaded pipelines beat streaming ones — and when they don't. Earlier, I built that instinct for rigor elsewhere — causal inference at Fidelity's AI Center of Excellence, low-latency concurrent systems at Density. Same lesson both times: a model is only as good as your ability to measure it. Dual degree (B.Tech + M.Tech), IIT Kharagpur — still the place that taught me to ask "why" before "how."

Location

Greater Bengaluru Area

AI stack

claude code

cursor

deepseek

elevenlabs

gemini

hugging face

Work experience

Artificial Intelligence Engineer at Sears, May 2025 - Present · 1 y 3 mo, Chicago, IL, Building Voice agents in retail space for the largest home services in the US

Data scientist intern at Fidelity Investments, May 2024 - Jul 2024 · 2 mo, Bengaluru, Developed and implemented causal inference methodologies to address uplift attribution challenges on financial datasets within the Financial Institution wing of Fidelity Investments' AI Center of Excellence. Designed Market Mix Models (MMM) and causal A/B testing framework improving marketing ROI attribution accuracy by 12%. • Enhanced MMM precision via Tweedie regression and temporal feature extraction (Prophet + TensorFlow),reducing MSE gap by 8%. • Enhanced KPI attribution techniques to zero-inflated panel data using Tweedie regressor models, increasing the R² coefficient by 40%.

Quantitative Research Analyst at Density, May 2023 - Aug 2023 · 3 mo, Bengaluru, Karnataka, India, •Co-developed a Python-based predictive trading framework using multi-processing for data ingestion and InfluxDB for trade logging. •Built ML models for trade signal generation using PCA on technical indicators and Monte Carlo simulations, achieving a 78% win rate. •Implemented backtesting with distribution modeling for slippage, simulating real-world market conditions and reducing trading costs by 15%. •Deployed the trading system on AWS, ensuring scalability and addressing challenges like drift with distributed computing (EC2, S3, Lambda). •Helped quant traders develop a price action-based strategy using statistical arbitrage

Education

Bachelors and Masters Integrated, Electrical, Electronics and Communications Engineering and Computer Vision, Indian Institute of Technology, Kharagpur

Ascent Classes

Public links

aicard-production.up.railway.app: https://aicard-production.up.railway.app/