Python / Network Engineer: Network Optimization of the Bot (WAF, C-lib integration, uvloop)
About the project: An automated bot for sniping. The basic business logic (parsing, authorization, assembling JSON/XML packets) is already ready and works stably on Python. The target environment is Ubuntu
Current problem: The architecture has hit the physical limits of the network. The combat cycle time is unsatisfactory due to network timeouts, aggressive limits of the platform's load balancer, and slow performance of the local cryptography module (Java). Our goal is to optimize the operation with http/1.1, bypass network delays, and squeeze the absolute maximum speed.
We are looking for an engineer who understands low-level network architecture, knows how WAFs (Rate Limits, Leaky Bucket) work, and can integrate system libraries into Python's memory.
Tasks (Development of an Isolated Network Engine):
Elimination of I/O delays (C/C++ Integration): Currently, cryptographic operations go through an external desktop Java application with data transfer via local sockets. Requirement: Completely eliminate this intermediary. Integrate a native system C library directly into Python's address space (via ctypes or CFFI), transferring all operations exclusively to RAM.
Bypassing WAF and warming up connections (Network Engine): The platform's protection cuts inactive TCP connections due to timeouts and bans for spam requests. Requirement: Write a network engine (based on uvloop) that preemptively opens a pool of TCP connections and keeps them "hot" (Keep-Alive) using a legal background "pulse," without attracting the attention of the protection system.
Monitoring floating start (Staggered Polling): The timing of the target event is unpredictable (floating start). Blindly spamming requests will lead to an IP ban. Requirement: Develop an asynchronous "radar" with phase offset (desynchronization of the pool) that allows monitoring the status of the lot with micro-delays, strictly adhering to the platform's limits (for example, 1 request per second per connection).
Session freezing (State Machine): Requirement: Implement strict sequential sending of multiple packets through a single warmed connection with dynamic updating of security tokens (CSRF/Sequence) to avoid session invalidation by the server under high loads.
Candidate requirements:
Deep understanding of asyncio and the structure of HTTP protocols.
Ability to analyze network traffic and bypass load balancer limits.
Experience with reverse engineering, parsing aggressively protected systems, HFT, or arbitrage scanners.
Skill in memory management and integrating C libraries in Linux.
Willingness to keep project details confidential (NDA).
Important: We need precise work with the network. Brute force and solutions at the level of while True: sleep() are not considered.
Project acceptance conditions (Critical): Payment for the project is strictly tied to Internal Overhead — the pure internal speed of your engine's execution. We need surgically optimized code without I/O blocking, unnecessary reconnections, and drops in the Event Loop. Exact acceptance metrics (acceptable milliseconds for each internal stage) and profiling/logging requirements will be provided to candidates with relevant experience in personal correspondence.
To respond: 1. Please briefly describe your experience in latency optimization or bypassing WAF. 2. Start your response with which library/approach you would use to solve task #3. 3. Propose realistic timelines and costs.
