Abhay

Key Skills

Software

Don't disclose

Other

Top Subject Matter

AI interview,voice agents, real-time call data for AI evaluation

Legal Services & Contract Review

Software engineer expertise in Python, TS, Go and Rust

Top Data Types

Text

Audio

Document

Top Task Types

Prompt + Response Writing (SFT)

Computer Programming/Coding

Evaluation/Rating

RLHF

Fine-tuning

Classification

Transcription

Function Calling

Freelancer Overview

Software Engineer | FullTime | HelloGenAI (AI interview agent and AI calling voice agent) (Nov'23 - Dec'24). Brings 4+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Don't disclose, Other, and Internal. Education includes Bachelor of Technology, Indian Institute of Technology Roorkee (2024) and Intermediate, Kamal Model International School (2020). AI-training focus includes data types such as Text and Audio and labeling workflows including Evaluation, Rating, and Transcription.

Labeling Experience

Founding Engineer | FullTime | Pocket (by SendAI) (Jan'25 - June'26)

TextFunction Calling

Designed agentic wallets and AI automation pipelines that transform natural-language inputs into blockchain tool/function calls. Implemented multi-agent workflow context management using RAG and MCP patterns to improve tool selection and reduce prompt bloat. Built reliability and observability features for mission-critical high-throughput on-chain and off-chain actions that depend on correct structured tool execution. • Integrated RAG pipelines and MCP patterns for maintaining context in multi-agent workflows. • Improved tool selection and reduced prompt bloat for more consistent function/tool calling. • Contributed Rust contract logic and Python/TypeScript orchestration backends to support reliable agent actions. • Added indexing and observability/scalability mechanisms for high-throughput execution.

2025 - 2026

Founding Engineer - Pocket

Audio

Built agentic wallets and AI automation pipelines that execute blockchain actions such as token transfers, staking, NFT minting, and DeFi interactions based on natural-language inputs. Designed scalable indexing for Solana using RPCs, WebSockets, and custom programs and implemented a DEX aggregator to source liquidity across protocols. Led integration of Retrieval-Augmented Generation (RAG) pipelines and Model Context Protocol (MCP) patterns to manage context across multi-agent workflows and reduce prompt bloat. • Engineered Rust smart-contract logic and Python/TypeScript orchestration backends • Implemented reliability, observability, and scalability practices for high-throughput on-chain and off-chain systems • Built blockchain data pipelines and indexing services • Improved tool selection and context management for multi-agent automation

2025 - 2026

Software Engineer | FullTime | HelloGenAI (AI interview agent and AI calling voice agent) (Nov'23 - Dec'24)

Don't discloseText

Built and maintained AI interview and AI voice-calling products serving large concurrent user traffic, which required continuous model behavior tuning and evaluation. Worked on systems that likely generated training/evaluation datasets from real call/interview transcripts and interaction logs. Ensured reliability and scalability so generated outputs could be assessed at production scale. • Developed an AI interview agent and an AI calling voice agent supporting 100K+ concurrent calls. • Designed analytics workflows using high-frequency call data streaming to support operational assessment. • Improved system performance to enable consistent generation and downstream evaluation. • Applied scalable AWS architecture for high availability and fault tolerance during evaluation runs.

2023 - 2024

Software Engineer (HelloGenAI) — AI interview agent and AI calling voice agent

Text

Built and maintained AI interview and voice agents intended to support large-scale, high-throughput conversations and call flows. Designed system components that enable downstream AI behavior tuning and quality evaluation by structuring real-time interaction data. Delivered production-ready AWS-based architecture to handle conversational workloads at scale. • AI interview agent development and operational support • AI calling voice agent development for concurrent calls • System architecture and real-time data streaming for analytics • High availability and fault tolerance for mission-critical AI services

2023 - 2024

MERN Developer | Freelance | Rhythmicstan Pvt Ltd (MockGPT project implied) (Mar'23 - Jun'23)

OtherAudioTranscription

Developed an AI-based mock interview platform that uses speech-to-text and text-to-speech to simulate realistic interview scenarios. Implemented bidirectional real-time communication to capture audio/transcripts for the interactive interview flow. Although not a human annotation workflow, the project required preparing and processing conversational audio data for model-driven outputs. • Integrated OpenAI, Deepgram (STT), and AWS Polly (TTS) for interview conversation. • Implemented WebSockets for real-time, bidirectional client-server communication. • Built a code editor for participants to write, submit, and have code evaluated within interviews. • Supported end-to-end transformation from audio to transcripts and back to synthesized speech.

2023 - 2023

Education

Work History