Akul Datta - Resume (Engineer & Researcher)

Professional Summary

Engineer & Researcher with an M.S. in Computer Science from the University of Illinois at Urbana-Champaign, specializing in LLMs, Computer Vision, and Robotics with a focus on building scalable, production-grade intelligent systems.

Professional Experience

Applied AI Engineer

Mercor — San Francisco, CA

Aug 2025 – Present

Published Mercor’s core AI benchmarks including APEX-SWE, APEX-v1, ACE, and APEX-Agents, testing complex agentic systems on viable economic workflows.
Led design and development of robust Reinforcement Learning (RL) sandbox environments optimized for code generation, safety layers, BUA/CUA compliance, and professional business domains.
Ran large-scale internal model post-training experiments on alignment datasets and spearheaded research on advanced LLM searching/planning structures.
Successfully scaled four critical data pipelines from scratch to generate multi-million dollar monthly revenue streams.

Core Contributor

ARC Prize — San Francisco, CA

Jun 2025 – Jul 2025

Engineered the foundational developer repository featuring training templates for the ARC-AGI-3 benchmark.
Collaborated with engineering teams from HuggingFace, Anthropic, and LangChain to integrate robust agent workflows.
Authored extensive developer documentation for the official global AI competition.

AI Engineering Intern

Endor Labs — Palo Alto, CA

May 2024 – Aug 2024

Architected an advanced human-in-the-loop LLM Security Agent to autonomously audit, optimize, and organize enterprise AppSec workflows.
Built an automated vulnerability classifier providing precise pull request SAST static analysis feedback.

AI Researcher

University of Texas at Austin

May 2023 – Dec 2023

Synthesized high-fidelity 3D datasets using the MuJoCo physics environment to significantly enhance out-of-distribution (OOD) computer vision bounding box detection.
Fine-tuned generative Stable Diffusion architectures, capturing state-of-the-art accuracy benchmarks on visual OOD validation tests.
Optimized large-scale distributed GPU cluster workflows for efficient multi-node model pre-training.

Undergraduate Researcher

University of Illinois at Urbana-Champaign

May 2023 – Dec 2023

Designed and implemented a novel State-of-the-Art (SoTA) Priority-Aware Load Balancer and Adaptive Concurrency Filter.
Greatly minimized high-traffic bottlenecks by separating network payloads into discrete priority paths using a custom queue filter.

Software Engineering Intern

Axiamatic — San Francisco, CA

May 2022 – Aug 2022

Developed a scalable, fault-tolerant enterprise distributed audit logger resolving 1,000+ events/sec at 99.99% system uptime.

Education

M.S. in Computer Science

University of Illinois at Urbana-Champaign

Aug 2024 – May 2025

Focus: Artificial Intelligence, Computer Vision, Drone Robotics, Control Algorithms.
Research Thesis: Neural networks optimized for occlusion-aware object tracking in autonomous drones.
Advisor: Prof. Sayan Mitra | Leadership: Team Lead - UIUC Drone Racing Team.

B.S. in Computer Science

University of Illinois at Urbana-Champaign

Aug 2021 – May 2024

Focus Area: Artificial Intelligence & Systems Programming.
Selected Coursework: Reinforcement Learning, Distributed Systems, Deep Learning, System Programming, Applied Machine Learning, Algorithm Design.

Peer-Reviewed Publications

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model

Acikgoz, E. C., Datta, A., et al.

Association for Computational Linguistics (ACL 2025 Main)

📄 Read Paper (OpenReview) →

From Turns to Dialogues: Rethinking Task-Oriented Dialogue Evaluation by Combining Turn-Level Precision with Dialogue-Level Comparisons

Datta, A., et al.

Special Interest Group on Discourse and Dialogue (SIGDial 2025)

📄 Read Paper (OpenReview) →

Co-Design of Communication and Machine Inference for Cloud Robotics

Nakanoya, M., Datta, A., et al.

Robotics: Science and Systems (RSS 2021)

📄 Read Paper (Proceedings) →

Technical Expertise

Languages & Cores

Python C++ Java CUDA SQL

AI / ML Frameworks

PyTorch TensorFlow Stable Diffusion OpenCV

Agentic Frameworks

LangGraph Anthropic MCP LangChain HuggingFace

Cloud & Robotics

AWS Azure GCP MuJoCo Docker

Awards & Recognition

🏆

3rd Place, AGI House AI Hackathon (June 2023) for developing a custom healthcare agent.

📜

California State Senate Commendation (2021) for leadership on public health advocacy.

💻

2nd Place, SIGPwny CTF CSAW'21 cybersecurity capture-the-flag tournament.

👑

Dean's List Award, UIUC College of Engineering (All semesters 2021 - 2023).

✨

3rd Place, PalyHacks Hackathon for custom generative ambient audio filter model.

📡

Founder of akul.ai, a high-growth AI newsletter followed by VCs, researchers, and engineers.