Shanto is available for hire
Hire ShantoShanto Mathew
Verified Expert in Engineering
Artificial Intelligence Engineer and Developer
Dallas, TX, United States
Toptal member since June 1, 2026
Bio
Shanto is an AI engineer specializing in agentic and voice AI. He builds production agents using Claude Code, Codex, and custom multi-agent frameworks, as well as low-latency voice agents for retail, support, and sales. Shanto takes frontier models from prototype to reliable, cost-aware production systems—tool use, RAG, orchestration, evaluations, and the integrations that make them real.
Portfolio
CDW
Python 3, Python, Python API, Cortex XSOAR, Cortex XSIAM, AWS IoT, Snowflake...
Independent Consulting
AI Agents, AI Voice Agents, Automation, Claude Code, Codex, Twilio, ElevenLabs...
Bank of America
LangGraph, LangChain, RAG Systems, Python 3, Splunk SOAR, AI Agents, FAISS...
Experience
- Automation - 8 years
- AI Agents - 7 years
- Python 3 - 7 years
- LangChain - 6 years
- AI Voice Agents - 5 years
- LangGraph - 2 years
- Codex - 1 year
- Claude Code - 1 year
Preferred Environment
PyCharm
The most amazing...
...thing I've built is a multi-agent AI that runs revenue operations: finds signals, scores accounts, drafts outreach, and writes CRM—human-approved.
Work Experience
Senior SOAR Engineer – Generative AI Security Automation
2026 - PRESENT
CDW
- Architected Cortex XSIAM and XSOAR playbooks with Python custom integrations for autonomous threat triage, alert summarization, and IOC extraction across multi-cloud environments.
- Implemented MCP patterns for context-aware incident correlation across multi-cloud environments, enabling intelligent automated remediation.
- Designed scalable Python data pipelines on AWS and Snowflake for security telemetry ingestion and analysis, supporting Fortune 500 SOC operations.
- Engineered prompt templates and LLM integrations within Palo Alto XSIAM for automated phishing analysis, malware classification, and incident summarization.
Technologies: Python 3, Python, Python API, Cortex XSOAR, Cortex XSIAM, AWS IoT, Snowflake, Model Context Protocol (MCP), AI Agents, Automation, AWS Lambda, REST APIs, API Design, Agentic Frameworks, Testing, Node.js, Event-driven Architecture, DevOps, JavaScript, Next.js, APIs, Databases, Generative Artificial Intelligence (GenAI), Minimum Viable Product (MVP), Product Management, AI Automation, LangSmith, Observability, Azure OpenAI Service, AI Agent Orchestration, Workflow, Product Frameworks
Senior Generative AI Engineer – Voice AI & Agentic AI
2025 - PRESENT
Independent Consulting
- Built Retell Copilot, a voice agent that provisions other Retell voice agents end-to-end in under 60 seconds from a caller's spec; deployed on AWS Lambda, API Gateway, DynamoDB, and Netlify edge.
- Shipped 7 live voice AI demos on Retell AI, OpenAI Realtime, xAI Grok Voice Agent, ElevenLabs, and Twilio across sales, medical, hotel, and enterprise concierge use cases.
- Architected real-time voice agents on browser WebSocket transport with streaming PCM audio, achieving sub-250 milliseconds time-to-first-audio through buffer tuning, prompt caching, and connection pooling.
- Built role-aware RAG pipelines with LangChain and LlamaIndex using semantic chunking, hybrid search, citation enforcement, and versioned prompt templates with an offline eval harness.
- Built LoRA and QLoRA fine-tuning pipelines using HuggingFace Transformers and custom training loops for cybersecurity and customer-service domains, improving task-specific accuracy by 15-30%.
- Built NetDebt AI Voice Agent for Landmark Management Group: live Retell/Twilio intake agent with 6 custom function-calling tools, pre-seeded lead database, and CloudFront-hosted dashboard.
- Implemented production-grade security boundaries for voice AI: server-side ephemeral client-secret minting, CSP and HSTS hardening, secret-safe deploys, and graceful 503 fallbacks when keys are absent.
- Achieved provider portability across Retell AI, OpenAI Realtime, xAI Grok Voice Agent, Anthropic Claude, and Z.ai GLM-5.1 through clean abstraction layers and an offline eval harness AB-tested in production.
Technologies: AI Agents, AI Voice Agents, Automation, Claude Code, Codex, Twilio, ElevenLabs, FastAPI, AWS Lambda, Amazon DynamoDB, Retrieval-augmented Generation (RAG), FAISS, Chroma, LoRa, QLoRA, LangChain, LangGraph, Python, Large Language Models (LLMs), Claude, Artificial Intelligence (AI), Architecture, Retell AI, Vapi, Customer Relationship Management (CRM), Real Estate, API Design, Agentic Frameworks, Error Handling, Azure, Open Source, Remote Work, Technical Leadership, Full-stack Development, Agentic Workflow Design, Node.js, PostgreSQL, Agentic AI, Agentic AI Systems, Distributed Systems, Large Language Model Operations (LLMOps), Event-driven Architecture, Event-driven Systems, Machine Learning Operations (MLOps), DevOps, Terraform, SaaS, Whisper, API Integration, Speech-to-Text (STT), AI Chatbots, Rust, AI Tools, Data Visualization, Firebase, RAG Architecture, AI Integration, APIs, Databases, Generative Artificial Intelligence (GenAI), Minimum Viable Product (MVP), Product Management, Containerization, Docker, Leadership, CSS, Front-end, React Router, Vite, Workflow Automation & System Integration, AI Automation, AI Architecture, Agentic RAG Systems, RAG Pipelines, AI Agent Orchestration, Workflow, Multi-agent Systems, Product Frameworks, OpenAI
Senior SOAR Engineer – AI Security Automation
2024 - 2025
Bank of America
- Developed 15 Splunk SOAR applications in Python with AI-enhanced custom functions for automated incident response workflows across the global SOC.
- Built LangGraph multi-agent SOAR proof-of-concept for Splunk ES notables (triage, enrich, correlate, decide) with MCP-guarded actions and integrations to Microsoft 365 Defender, CrowdStrike, and ServiceNow.
- Shipped production agentic AI on LangGraph, automating L1/L2 SOC workflows, achieving 60% reduction in manual analyst intervention.
- Engineered RAG pipelines with ChromaDB and FAISS over MITRE ATT&CK and threat intel, enabling agents to retrieve TTP context for notable enrichment.
- Partnered with SOC leadership to define KPIs for AI-assisted triage, reducing mean time to investigate by 45% across high-volume notable categories.
Technologies: LangGraph, LangChain, RAG Systems, Python 3, Splunk SOAR, AI Agents, FAISS, ChromaDB, MITRE ATT&CK, Prompt Engineering, Large Language Models (LLMs), Claude, Artificial Intelligence (AI), REST APIs, ElevenLabs Solutions, Retell AI, Vapi, Compass, Customer Relationship Management (CRM), Real Estate, Error Handling, Remote Work, Full-stack, Full-stack Development, Agentic Workflow Design, PostgreSQL, Agentic AI, Express.js, Agentic AI Systems, Event-driven Systems, SaaS, Whisper, Speech-to-Text (STT), Terraform, Rust, AI Tools, Data Visualization, Firebase, RAG Architecture, AI Integration, JavaScript, Vanilla JS, APIs, Databases, Minimum Viable Product (MVP), CSS, React Router, Vite, Workflow Automation & System Integration, AI Automation, AI Architecture, Multi-agent Systems, OpenAI
Senior AI & Voice AI Engineer
2022 - 2024
CDW
- Built production Voice AI agents with Retell AI and OpenAI for outbound customer engagement, handling 10,000+ calls/month with sub-300-millisecond response latency.
- Designed LangGraph-based conversation flows with dynamic function calling, enabling agents to book appointments, qualify leads, and escalate to humans.
- Integrated Twilio telephony, WebSockets, and streaming TTS/STT pipelines, achieving natural barge-in and interruption handling across voice sessions.
- Implemented evaluation harnesses, call analytics, and prompt tuning loops that improved task completion rate by 35% across voice agent campaigns.
- Architected secure XSIAM/XSOAR Python automations integrating 30+ enterprise security tools, reducing alert fatigue for Fortune 500 SOC clients.
Technologies: Automation, AI Agents, AI Voice Agents, AWS IoT, AWS Lambda, Chroma, Twilio, Large Language Models (LLMs), Conversational AI, Prompt Engineering, WebSockets, Architecture, Compass, Open Source, Testing, Agentic Workflow Design, Agentic AI Systems, Machine Learning Operations (MLOps), AI Chatbots, AI Tools, AI Architecture, OpenAI
Senior Machine Learning Engineer
2020 - 2023
Mastercard
- Developed ML models in Python and Spark to detect fraudulent transactions across millions of daily payments, improving precision by 22% over baseline.
- Built end-to-end MLOps pipelines for model training, validation, deployment, and monitoring on AWS, reducing model release cycle from weeks to days.
- Engineered large-scale feature pipelines on Spark and SQL over Mastercard transaction data, powering risk scoring and customer segmentation models.
- Collaborated with risk and product teams to translate domain requirements into ML solutions, presenting model results and trade-offs to executive stakeholders.
Technologies: SQL, Python, Machine Learning, Azure, Remote Work
Experience
Agentic AI SOC Copilot – LangGraph + RAG
Designed and built a multi-agent SOC copilot using LangGraph that ingests Splunk ES notables and orchestrates specialized agents for triage, enrichment, correlation, and decisioning.
The system uses a RAG layer over MITRE ATT&CK, internal runbooks, and historical incidents (ChromaDB + FAISS) so agents can ground responses in real TTP context. MCP-style tool guardrails control actions across Microsoft 365 Defender, CrowdStrike, and ServiceNow, ensuring every action is policy-checked and auditable.
I implemented prompt engineering patterns, structured outputs, evaluation harnesses, and observability hooks to enable analysts to review and override agent decisions. The deployment reduced manual L1/L2 analyst intervention by 60% and cut mean time to investigate by 45% on high-volume notable categories. Built with Python 3, LangGraph, LangChain, OpenAI, ChromaDB, FAISS, Splunk SOAR, and FastAPI.
The system uses a RAG layer over MITRE ATT&CK, internal runbooks, and historical incidents (ChromaDB + FAISS) so agents can ground responses in real TTP context. MCP-style tool guardrails control actions across Microsoft 365 Defender, CrowdStrike, and ServiceNow, ensuring every action is policy-checked and auditable.
I implemented prompt engineering patterns, structured outputs, evaluation harnesses, and observability hooks to enable analysts to review and override agent decisions. The deployment reduced manual L1/L2 analyst intervention by 60% and cut mean time to investigate by 45% on high-volume notable categories. Built with Python 3, LangGraph, LangChain, OpenAI, ChromaDB, FAISS, Splunk SOAR, and FastAPI.
Agentic Marketing Operations Workbench
https://gp-agentic-revenue-ops.netlify.app/Built an agentic revenue-ops workbench that streams live buyer-intent signals from public sources via GLM-5.1 + web_search, scoring each account against ICP fit and brand-voice thresholds in real time.
I designed a signals dashboard tracking hiring, funding, expansion, exec-hire, and compliance triggers across global accounts, with approval queues, in-flight run telemetry (p95 22.4s), and live streaming refresh. I implemented multi-agent orchestration with degraded-mode fallbacks, connector health monitoring, and human-in-the-loop approvals before any outbound action. I delivered a production-grade UI with operator-friendly filters, regional segmentation, and audit-ready signal provenance for go-to-market teams.
I designed a signals dashboard tracking hiring, funding, expansion, exec-hire, and compliance triggers across global accounts, with approval queues, in-flight run telemetry (p95 22.4s), and live streaming refresh. I implemented multi-agent orchestration with degraded-mode fallbacks, connector health monitoring, and human-in-the-loop approvals before any outbound action. I delivered a production-grade UI with operator-friendly filters, regional segmentation, and audit-ready signal provenance for go-to-market teams.
Enterprise Voice AI Launch Console (VoiceOps)
https://elevenlabs-forward-deployed-engineer.netlify.app/Designed a forward-deployed cockpit for shipping ElevenAgents' voice AI from enterprise discovery into production.
I built launch-room, architecture, simulation, safety, stakeholder, and productize modules covering API integrations, evaluation results, executive updates, and reusable pilot kits. I also implemented launch-gate checks, synthetic-data evaluations, and hold-launch controls to capture risk before go-live, plus repeatable customer feedback loops for an enterprise voice rollout. I shipped a polished operator UI that mirrors how a forward-deployed engineer owns architecture, runs go-live reviews, and turns one customer launch into a productized motion across accounts.
I built launch-room, architecture, simulation, safety, stakeholder, and productize modules covering API integrations, evaluation results, executive updates, and reusable pilot kits. I also implemented launch-gate checks, synthetic-data evaluations, and hold-launch controls to capture risk before go-live, plus repeatable customer feedback loops for an enterprise voice rollout. I shipped a polished operator UI that mirrors how a forward-deployed engineer owns architecture, runs go-live reviews, and turns one customer launch into a productized motion across accounts.
Grok Voice Medical Front Desk
https://grok-medical-frontdesk.netlify.app/Built a live voice-controlled clinic front desk powered by xAI Grok, where a synthetic receptionist handles appointment scheduling, availability checks, plan verification, and office FAQs over real-time audio.
I implemented streaming caller mic and receptionist audio with live transcript, caller intake context, open-slot inventory, and console navigation by voice. I enforced no-medical-advice safety boundaries, PHI-free logging, and an automatic 5-minute call disconnect to satisfy compliance guardrails. I delivered Front Desk, Schedule, and Audit views, plus a synthetic demo mode, so prospects and operators can exercise the agent without exposing real patient data.
I implemented streaming caller mic and receptionist audio with live transcript, caller intake context, open-slot inventory, and console navigation by voice. I enforced no-medical-advice safety boundaries, PHI-free logging, and an automatic 5-minute call disconnect to satisfy compliance guardrails. I delivered Front Desk, Schedule, and Audit views, plus a synthetic demo mode, so prospects and operators can exercise the agent without exposing real patient data.
Y22 Voice AI Sales Roleplay Simulator
https://y22-ai-sales-roleplay.netlify.app/Designed Y22 Roleplay, a voice-driven AI sales training simulator where reps practice live calls against synthetic buyer personas powered by a Grok voice think-fast model.
I built 3 preset buyer archetypes (Skeptical Mid-Market CFO, Friendly-but-Firm VP Sales, Procurement Gatekeeper) with difficulty tiering, plus a custom buyer builder so managers can spin up any persona on demand. I implemented a six-tile behavior scorecard that fills in live during the call, a Prompt Lab for tuning persona prompts, and an end-of-call scorecard to coach reps. I streamed real-time voice with synthetic-data-only safety mode, making it safe to ship to enterprise sales organizations as a daily practice tool.
I built 3 preset buyer archetypes (Skeptical Mid-Market CFO, Friendly-but-Firm VP Sales, Procurement Gatekeeper) with difficulty tiering, plus a custom buyer builder so managers can spin up any persona on demand. I implemented a six-tile behavior scorecard that fills in live during the call, a Prompt Lab for tuning persona prompts, and an end-of-call scorecard to coach reps. I streamed real-time voice with synthetic-data-only safety mode, making it safe to ship to enterprise sales organizations as a daily practice tool.
SOC AI Agent – LangGraph Incident Investigator
https://security-ops-playbook-analyzer.netlify.app/Built a 1-button SOC investigation agent with visible graph mechanics, powered by a LangGraph incident-investigation pipeline on a live GLM-5.1 back end.
The agent streams graph execution, checkpoint snapshots, tool calls, and human approval interrupts while it triages a freshly generated incident—covering enrichment, log analysis, ticketing, and remediation steps. Surfaced operator-grade telemetry: checkpoints, tool-call counts, token usage, and MTTR per run, plus parallel Send() fan-out and map-reduce back-edges to demonstrate real LangGraph patterns. I packaged the demo as a click-to-run experience so SOC managers can see how agentic AI compresses Tier-1 triage without giving up auditability or approval gates.
The agent streams graph execution, checkpoint snapshots, tool calls, and human approval interrupts while it triages a freshly generated incident—covering enrichment, log analysis, ticketing, and remediation steps. Surfaced operator-grade telemetry: checkpoints, tool-call counts, token usage, and MTTR per run, plus parallel Send() fan-out and map-reduce back-edges to demonstrate real LangGraph patterns. I packaged the demo as a click-to-run experience so SOC managers can see how agentic AI compresses Tier-1 triage without giving up auditability or approval gates.
Education
2006 - 2010
Bachelor of Technology Degree in Computer Science
Mahatma Gandhi University - Kerala, India
Certifications
MAY 2026 - PRESENT
Claude Code Certification
Anthropic
Skills
Libraries/APIs
React, REST APIs, Node.js, React Router, Python API
Tools
Claude Code, Codex, Splunk SOAR, Claude, Retell AI, Terraform, Whisper, Azure OpenAI Service, PyCharm, Splunk
Languages
Python 3, SQL, Python, TypeScript, Rust, JavaScript, CSS, Snowflake
Frameworks
LangGraph, Compass, Agentic Frameworks, Express.js, Next.js, Vanilla JS
Paradigms
Automation, Testing, Event-driven Architecture, DevOps, Model Context Protocol (MCP)
Platforms
Twilio, Azure, Firebase, Docker, LangSmith, AWS Lambda, Cortex XSOAR, AWS IoT
Storage
Amazon DynamoDB, PostgreSQL, Databases
Other
AI Agents, AI Voice Agents, LangChain, RAG Systems, Retrieval-augmented Generation (RAG), Prompt Engineering, Large Language Models (LLMs), LLM Integration, ElevenLabs Solutions, Artificial Intelligence (AI), Architecture, Vapi, Customer Relationship Management (CRM), Real Estate, API Design, Error Handling, Open Source, Remote Work, Technical Leadership, Full-stack, Full-stack Development, Agentic Workflow Design, Agentic AI, Agentic AI Systems, Distributed Systems, Large Language Model Operations (LLMOps), Event-driven Systems, Machine Learning Operations (MLOps), SaaS, API Integration, Speech-to-Text (STT), AI Chatbots, AI Tools, Data Visualization, RAG Architecture, AI Integration, APIs, Generative Artificial Intelligence (GenAI), Minimum Viable Product (MVP), Product Management, Containerization, Leadership, Front-end, Vite, Workflow Automation & System Integration, AI Automation, Observability, AI Architecture, Agentic RAG Systems, RAG Pipelines, AI Agent Orchestration, Workflow, Multi-agent Systems, Product Frameworks, OpenAI, Splunk Enterprise Security, Professional Services, ElevenLabs, FastAPI, FAISS, Chroma, LoRa, QLoRA, Cortex XSIAM, ChromaDB, MITRE ATT&CK, Conversational AI, WebSockets, Machine Learning, Computer Science, Data Structures, Algorithms, Grok, Speech Recognition
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring