For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
D
Delmond Bongha

Delmond Bongha

Sales Playbook RAG Agent – Portfolio Project

Cameroon flagDouala, Cameroon
IntermediateOther

Key Skills

Software

Other

Top Subject Matter

Sales Playbook
Sales Knowledge Base
Sales Email Analysis

Top Data Types

TextText
AudioAudio
DocumentDocument

Top Task Types

Question AnsweringQuestion Answering
ClassificationClassification
TranscriptionTranscription

Freelancer Overview

Sales Playbook RAG Agent – Portfolio Project. Brings 4+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Other. Education includes Bachelor of Technology, University of Bamenda (2022) and Professional Certificate, Coursera / Google (2026). AI-training focus includes data types such as Text and Audio and labeling workflows including Question Answering, Classification, and Transcription.

Intermediate

Labeling Experience

Prospect Research Agent – Portfolio Project

OtherTextClassification
Designed an autonomous research agent that creates structured sales briefs for target companies by extracting and classifying data from web searches and LinkedIn profiles. Modelled all agent outputs as Pydantic schemas to ensure structured and validated results for CRM integration. Implemented caching and concurrency for efficient data collection and reduced redundancy. • Fetched and classified company intelligence via web scraping and aggregation tools. • Validated classification outputs with strict schema enforcement to prevent errors. • Cached previous results to optimize automated annotation processes. • Integrated agent outputs seamlessly into sales prospecting workflows.

Designed an autonomous research agent that creates structured sales briefs for target companies by extracting and classifying data from web searches and LinkedIn profiles. Modelled all agent outputs as Pydantic schemas to ensure structured and validated results for CRM integration. Implemented caching and concurrency for efficient data collection and reduced redundancy. • Fetched and classified company intelligence via web scraping and aggregation tools. • Validated classification outputs with strict schema enforcement to prevent errors. • Cached previous results to optimize automated annotation processes. • Integrated agent outputs seamlessly into sales prospecting workflows.

2025 - 2025

AI Sales Call Coach – Portfolio Project

OtherAudioTranscription
Built an end-to-end voice intelligence and transcription pipeline for sales call recordings. Used Whisper to transcribe audio and performed dimension-based call analysis using Claude API. Automated the generation and storage of structured coaching notes in Salesforce via API integration. • Transcribed audio with speaker diarisation for multi-speaker calls. • Automated scoring across 10 sales dimensions, delivering actionable insights for coaching. • Enabled natural language queries on historical call data through a custom MCP server. • Visualized performance trends and improvement curves on a web dashboard.

Built an end-to-end voice intelligence and transcription pipeline for sales call recordings. Used Whisper to transcribe audio and performed dimension-based call analysis using Claude API. Automated the generation and storage of structured coaching notes in Salesforce via API integration. • Transcribed audio with speaker diarisation for multi-speaker calls. • Automated scoring across 10 sales dimensions, delivering actionable insights for coaching. • Enabled natural language queries on historical call data through a custom MCP server. • Visualized performance trends and improvement curves on a web dashboard.

2025 - 2025

Sales Email Analyzer CLI – Portfolio Project

OtherTextClassification
Engineered a batch CLI tool that processes sales email threads via Claude API and extracts structured deal metadata. All LLM outputs are validated as Pydantic models and exported to CSV for CRM import. Enforced strict data schema standards to ensure zero hallucinated fields in classification results. • Automated extraction and classification of sentiment, action items, priority, and follow-up recommendations. • Achieved fast throughput by concurrent API calls, processing 50 emails in 30 seconds. • Provided outputs as validated JSON models and CSV files for downstream automation. • Used Python, Anthropic API, and Pydantic for scalable annotation pipeline.

Engineered a batch CLI tool that processes sales email threads via Claude API and extracts structured deal metadata. All LLM outputs are validated as Pydantic models and exported to CSV for CRM import. Enforced strict data schema standards to ensure zero hallucinated fields in classification results. • Automated extraction and classification of sentiment, action items, priority, and follow-up recommendations. • Achieved fast throughput by concurrent API calls, processing 50 emails in 30 seconds. • Provided outputs as validated JSON models and CSV files for downstream automation. • Used Python, Anthropic API, and Pydantic for scalable annotation pipeline.

2025 - 2025

Sales Playbook RAG Agent – Portfolio Project

OtherTextQuestion Answering
Built a production RAG system ingesting PDFs, DOCX, call transcripts, and Notion exports into a Pinecone vector index for question-answering tasks. Developed and deployed automated evaluation suites for tracking faithfulness, context recall, and answer relevance using RAGAS. Enabled natural language querying with source-cited answers through a Gradio chat UI and Slack REST integration. • Conducted hybrid search (dense + sparse) using Cohere Rerank for improved answer selection. • Maintained high answer relevance (94%) based on internal evaluation sets. • Automated evaluation and reporting of model outputs in production setups. • Handled document ingestion, context extraction, and quality control for large knowledge bases.

Built a production RAG system ingesting PDFs, DOCX, call transcripts, and Notion exports into a Pinecone vector index for question-answering tasks. Developed and deployed automated evaluation suites for tracking faithfulness, context recall, and answer relevance using RAGAS. Enabled natural language querying with source-cited answers through a Gradio chat UI and Slack REST integration. • Conducted hybrid search (dense + sparse) using Cohere Rerank for improved answer selection. • Maintained high answer relevance (94%) based on internal evaluation sets. • Automated evaluation and reporting of model outputs in production setups. • Handled document ingestion, context extraction, and quality control for large knowledge bases.

2025 - 2025

Education

C

Coursera / Google

Professional Certificate, Artificial Intelligence

Professional Certificate
2026 - 2026
U

University of Bamenda

Bachelor of Technology, Software Engineering

Bachelor of Technology
2018 - 2022

Work History

I

Independent / Freelance

Backend Developer

N/A
2023 - Present