Osama Altaf

Software Engineer

Pakistan1 yr 1 mo experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in building scalable LLM-powered AI systems.
Proven track record in real-time multilingual communication.
Strong background in voice AI and multi-agent orchestration.

Stackforce AI infers this person is a SaaS-focused AI Engineer with expertise in voice and language technologies.

Contact

Skills

Core Skills

Large Language Model Operations (llmops)Artificial Intelligence (ai)Ai System ArchitectureVoice Ai AgentsVoice Ai & Conversational AiReal-time Communication SystemsMachine Learning

Other Skills

LangGraphPyTorchMLOpsMulti-Turn Dialogue SystemsLLM IntegrationSpeech-to-Text (STT) & Text-to-Speech (TTS)STT & TTS Fine_TuningPythonData PreprocessingModel TrainingML PipelinesWhisperMultilingual AI SystemsLow-Latency SystemsLarge Language Models (LLMs)

About

I’m an AI Engineer focused on building real-world, production-grade AI systems — especially LLM-powered, voice-enabled, and agentic platforms. My work spans multilingual voice AI, real-time translation systems, telephony agents, and large-scale LLM deployments. I’ve contributed to government-grade and enterprise projects, including a real-time communication system supporting 30+ languages with sub-1500ms latency. Technically, I work across the full stack of modern AI systems: • LLM inference and optimization using vLLM (KV cache, quantization) and AIBrix • Multi-agent orchestration with CrewAI, LangGraph, and n8n. • Voice AI pipelines using Whisper, LiveKit, Pipecat, STT/TTS systems • LLM fine-tuning and alignment on proprietary datasets • Scalable APIs with FastAPI, Redis, and production caching layers I enjoy turning research into scalable systems — optimizing models for speed, cost, and reliability in real deployments. Currently open to: AI Engineer | LLM Engineer | Voice AI Engineer | Agentic AI Roles (Remote / Global) 📩 Email: osamaaltaf.pk@gmail.com

Experience

Quickcall

AI Engineer – LLM System Research & Analyst (Contract)

Jun 2025 – Jan 2026 · 7 mos · Miami, Florida, United States · Remote

Working on company-level initiatives to deploy and scale large language models for smart assistants and task automation.
Implementing high-throughput LLM inference using vLLM with efficient KV cache utilization.
Integrating AIBrix (TikTok’s multi-agent orchestration framework) to coordinate tools and APIs in goal-driven agent workflows.
Fine-tuning LLMs on proprietary datasets to align outputs with domain-specific terminology.
Building scalable backend APIs using FastAPI and Redis-based in-memory caching (LMCache).

Large Language Model Operations (LLMOps)Artificial Intelligence (AI)

Fiverr, freelancer

AI Engineer – Voice AI Agent Platform

Feb 2025 – Nov 2025 · 9 mos · Miami, Florida, United States · Remote

Designed a scalable Voice AI platform supporting inbound and outbound telephony via SIP routing.
Built modular, no-code agent workflow builders for dynamic call flows.
Integrated LangGraph-powered LLM agents for multi-turn, context-aware conversations.
Collaborated on end-to-end STT and TTS pipeline architecture.

AI System ArchitectureLangGraphVoice AI Agents

Confidential govtech project

AI Voice Agent Developer & Research Analyst

Feb 2025 – Jul 2025 · 5 mos · Jordan · Remote

Co-developed a government-grade real-time multilingual translation system supporting 30+ languages.
Integrated Whisper ASR with custom speech pipelines achieving sub-1500ms latency.
Built real-time voice communication using LiveKit and Pipecat.
Designed context-aware memory modules to preserve conversation meaning across languages.
Ensured compliance with strict security, privacy, and performance standards.

Voice AI & Conversational AIReal-Time Communication Systems

Iaxon

Junior AI Engineer

Jul 2023 – Aug 2024 · 1 yr 1 mo · Pakistan · On-site

Worked on machine learning model development and experimentation for client-facing projects.
Built and maintained Python-based ML pipelines for data preprocessing, training, and evaluation.
Fine-tuned machine learning models to improve performance and adaptability across different use cases.
Assisted in deploying ML components into backend systems and internal tools.
Automated repetitive data processing tasks to support faster model iteration.
Collaborated with senior engineers to debug, optimize, and improve existing ML workflows.

Machine LearningPyTorch