Kushagra Deep profile photo 1 on Tal

Kushagra Deep

GenAI

Bengaluru South

8+ years of work experience

SimpliAI
Feb 20242 yrs 8 mosPresent

Bengaluru, Karnataka, India

-Built end to end conversational Agents capable of booking appointments , question answering , drop chase sales -Built low latency , low cost ,high accuracy RAG workflows -Built Agentic Workflows using Langgraph with different nodes catering to different customer query scenarios -Explored and Benchmarked over 50 open Source LLM’s for performance and Latency. -Prompt Engineering for multifarious production ready scenarios like Appointment Booking , Drop chase Lead Sale Assist, Loyalty Points Conversion. -Explored multiple open source LLM , TTS , ASR and benchmarked them for accuracy and latency
Times Internet
Jun 20212 yrs 5 mosOct 2023

Bengaluru, Karnataka, India

- Ideated and developed a contextual advertisement recommendation engine -Developed a fully functional social media virality prediction solution using Bi Directional LSTM. -Gender Prediction using Locality Sensitive Hashing. -ERPM estimator using Deep Learning in TensorFlow2.0 -Model Optimizations using Quantization Aware Training and Pruning -Regression using 2D CNN -POC’s : DeepFM, TFT
Mobileum
May 20192 yrsApr 2021

Gurgaon, Haryana, India

-Designed TripMetrics for Roaming Telecom users streaming data . This trip metic captured mobility and usage events of the mobile user. These data points were divided in four key services: Registration, SMS,Voice and Data. Each service was given a service score based on the quality of service discerned from data received. Individual scores were aggregated to give RoamingCustomerExperience Score(R-CX). -Designed and implemented Bloom Filters for faster lookups into BigData. -Correlated two streams of data using Spark Structured Streaming, Stateful spark API's. - Designed and Developed Realtime Data And KPI Alerting Pipeline using Kafka -Tech Stack: Hadoop,SparkSql,Hive,Hbase,Presto,OracleRDBMS,Structured Streaming,Java8
Next Education India Pvt Ltd
Jun 20161 yr 9 mosFeb 2018

Hyderabad, Telangana, India

- Worked in Microservices Architecture using Netflix API's : Zuul,Ribbon,Hysterix,Eureka. - RDBMS,SQL,Java7

Kushagra’s AI Stack

claude code
claude code
elevenlabs
elevenlabs
gemini
gemini
Cerebras
Cerebras
ChatGPT
ChatGPT
Deepgram
Deepgram

Education

Indian Institute of Technology, Delhi

Bachelor of Technology - BTech

Indian Institute of Technology, Delhi

Join Kushagra and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Kushagra Deep on Tal

Headline

GenAI

About

AI professional with hands on experience in building scalable, cost effective AI agents from scratch using SOTA LLM’s , TTS, STT. Looking for Gen AI roles only , more in terms of partnerships than employment

Location

Bengaluru South

AI stack

claude code

elevenlabs

gemini

Cerebras

ChatGPT

Deepgram

Work experience

Founding Member - Head of AI at SimpliAI, Feb 2024 - Present · 2 yrs 8 mos, Bengaluru, Karnataka, India, -Built end to end conversational Agents capable of booking appointments , question answering , drop chase sales -Built low latency , low cost ,high accuracy RAG workflows -Built Agentic Workflows using Langgraph with different nodes catering to different customer query scenarios -Explored and Benchmarked over 50 open Source LLM’s for performance and Latency. -Prompt Engineering for multifarious production ready scenarios like Appointment Booking , Drop chase Lead Sale Assist, Loyalty Points Conversion. -Explored multiple open source LLM , TTS , ASR and benchmarked them for accuracy and latency

AI Engineer at Times Internet, Jun 2021 - Oct 2023 · 2 yrs 5 mos, Bengaluru, Karnataka, India, - Ideated and developed a contextual advertisement recommendation engine -Developed a fully functional social media virality prediction solution using Bi Directional LSTM. -Gender Prediction using Locality Sensitive Hashing. -ERPM estimator using Deep Learning in TensorFlow2.0 -Model Optimizations using Quantization Aware Training and Pruning -Regression using 2D CNN -POC’s : DeepFM, TFT

AI - Data Science and Engineering at Mobileum, May 2019 - Apr 2021 · 2 yrs, Gurgaon, Haryana, India, -Designed TripMetrics for Roaming Telecom users streaming data . This trip metic captured mobility and usage events of the mobile user. These data points were divided in four key services: Registration, SMS,Voice and Data. Each service was given a service score based on the quality of service discerned from data received. Individual scores were aggregated to give RoamingCustomerExperience Score(R-CX). -Designed and implemented Bloom Filters for faster lookups into BigData. -Correlated two streams of data using Spark Structured Streaming, Stateful spark API's. - Designed and Developed Realtime Data And KPI Alerting Pipeline using Kafka -Tech Stack: Hadoop,SparkSql,Hive,Hbase,Presto,OracleRDBMS,Structured Streaming,Java8

R&D Software Engineer at Next Education India Pvt Ltd, Jun 2016 - Feb 2018 · 1 yr 9 mos, Hyderabad, Telangana, India, - Worked in Microservices Architecture using Netflix API's : Zuul,Ribbon,Hysterix,Eureka. - RDBMS,SQL,Java7

Education

Bachelor of Technology - BTech, Indian Institute of Technology, Delhi